Discussion:
New file uploaded to hercules-390
h***@public.gmane.org
2010-09-06 13:57:16 UTC
Permalink
Hello,

This email message is a notification to let you know that
a file has been uploaded to the Files area of the hercules-390
group.

File : /speed390.zip
Uploaded by : kerravon86 <kerravon86-/***@public.gmane.org>
Description : speed test of GCCMVS on PDOS/390

You can access this file at the URL:
http://groups.yahoo.com/group/hercules-390/files/speed390.zip

To learn more about file sharing for your group, please visit:
http://help.yahoo.com/l/us/yahoo/groups/original/members/forms/general.htmlfiles

Regards,

kerravon86 <kerravon86-/***@public.gmane.org>
kerravon86
2010-09-06 14:03:30 UTC
Permalink
I have just uploaded this:

http://groups.yahoo.com/group/hercules-390/files/speed390.zip

which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.

The compile is single-threaded, so it tests how
fast a single core is.

Here are the results so far:

1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz

It would be good if people could report speeds
on their computer.

Note that you can use it to test any version of
Hercules, basically. doit.bat is this simple:

set HERCULES_RC=pdos.rc
hercules -f pdos.cnf >hercules.log
set HERCULES_RC=

I'd like to buy a new computer sometime, and I'm
interested in a laptop actually worth buying
because I can see a difference.

Thanks. Paul.
kerravon86
2010-09-06 14:06:34 UTC
Permalink
Oh, one more thing I'm really interested in is
whether this application is memory-access-bound,
thus increasing the speed of memory is the thing
that makes the big difference, rather than the
CPU.

If anyone has the ability to change the speed
of their memory.

BFN. Paul.
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.
The compile is single-threaded, so it tests how
fast a single core is.
1:37 core i3 M330 2.13 GHz
It would be good if people could report speeds
on their computer.
Note that you can use it to test any version of
set HERCULES_RC=pdos.rc
hercules -f pdos.cnf >hercules.log
set HERCULES_RC=
I'd like to buy a new computer sometime, and I'm
interested in a laptop actually worth buying
because I can see a difference.
Thanks. Paul.
Joe Monk
2010-09-06 16:29:52 UTC
Permalink
IMHO this is a skewed test.



You are EMULATING the cpu. thus you are not testing the actual CPU speed.
You are testing how fast the real CPU can EMULATE the S390 CPU.



Joe



From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of kerravon86
Sent: Monday, September 06, 2010 9:04 AM
To: hercules-390-***@public.gmane.org
Subject: [hercules-390] speed test of GCCMVS on PDOS/390





I have just uploaded this:

http://groups.yahoo.com/group/hercules-390/files/speed390.zip

which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.

The compile is single-threaded, so it tests how
fast a single core is.

Here are the results so far:

1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz

It would be good if people could report speeds
on their computer.

Note that you can use it to test any version of
Hercules, basically. doit.bat is this simple:

set HERCULES_RC=pdos.rc
hercules -f pdos.cnf >hercules.log
set HERCULES_RC=

I'd like to buy a new computer sometime, and I'm
interested in a laptop actually worth buying
because I can see a difference.

Thanks. Paul.





[Non-text portions of this message have been removed]
kerravon86
2010-09-06 19:12:19 UTC
Permalink
Post by Joe Monk
IMHO this is a skewed test.
You are EMULATING the cpu. thus you are not testing
the actual CPU speed.
You are testing how fast the real CPU can EMULATE the S390 CPU.
Which happens to be exactly what Hercules users
are interested in? Well this one, at least!

When I upgrade my laptop, I want it to be optimized
for Hercules running in S/390 mode, as that is my
medium-term intended usage.

And as you noted, due to the unusual nature of the
work, it may be largely memory-bound (doing large
memcpy or memcmp operations), thus expensive CPUs
don't do much for performance. But I have no way
of finding out either way, as I'm not set up to
replace memory. I can at least measure what can be
measured though, which should be good enough for
a buying guide:

speed390:

1:13 AMD Phenom X4, rated at 2.6 GHz
1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz

BFN. Paul.
Joe Monk
2010-09-06 20:22:46 UTC
Permalink
Paul,



Right. BUT there are so many OTHER factors, that this test will not tell
you!



That's why it is a skewed test.



For instance, lets talk motherboard design. What CHIPSET? VIA, Intel? Maker?
Asus, Acer, Intel, Gigabyte?



Joe



From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of kerravon86
Sent: Monday, September 06, 2010 2:12 PM
To: hercules-390-***@public.gmane.org
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by Joe Monk
IMHO this is a skewed test.
You are EMULATING the cpu. thus you are not testing
the actual CPU speed.
You are testing how fast the real CPU can EMULATE the S390 CPU.
Which happens to be exactly what Hercules users
are interested in? Well this one, at least!

When I upgrade my laptop, I want it to be optimized
for Hercules running in S/390 mode, as that is my
medium-term intended usage.

And as you noted, due to the unusual nature of the
work, it may be largely memory-bound (doing large
memcpy or memcmp operations), thus expensive CPUs
don't do much for performance. But I have no way
of finding out either way, as I'm not set up to
replace memory. I can at least measure what can be
measured though, which should be good enough for
a buying guide:

speed390:

1:13 AMD Phenom X4, rated at 2.6 GHz
1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz

BFN. Paul.





[Non-text portions of this message have been removed]
kerravon86
2010-09-06 20:45:56 UTC
Permalink
Post by Joe Monk
Right. BUT there are so many OTHER factors, that this
test will not tell you!
That's why it is a skewed test.
For instance, lets talk motherboard design. What CHIPSET?
VIA, Intel? Maker?
Asus, Acer, Intel, Gigabyte?
I would have thought this was exactly what the
test is designed to show - does the motherboard
make a difference? Does the ram speed make a
difference? If so, how much?

That way people wondering whether to go for an
expensive motherboard, with a fantastic list of
specs from the manufacturer, can take 1 minute
and 3 seconds of their time to find out how the
manufacturer's claims stack up against a real
CPU-intensive, relevant task.

Can you predict the outcome? I can't. I haven't
run a profiler over it. One thing I know though
is that the MADDR routine is highly sensitive
to adding even a single test to it - sensitive
as in you'll add 3% by doing that. My guess is
that that is memory retrieval, which would
possibly account for the fact that the figures
we are getting are not dramatically different.
From what I remember of the 386 days, you'd
upgrade processor and get a 10-fold performance
boost.

BFN. Paul.
Joe Monk
2010-09-06 21:01:56 UTC
Permalink
Post by kerravon86
From what I remember of the 386 days, you'd
upgrade processor and get a 10-fold performance
boost.
Not anymore. You can upgrade processor, but that alone wont give you what
you are trying to achieve.



Joe



From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of kerravon86
Sent: Monday, September 06, 2010 3:46 PM
To: hercules-390-***@public.gmane.org
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by kerravon86
Right. BUT there are so many OTHER factors, that this
test will not tell you!
That's why it is a skewed test.
For instance, lets talk motherboard design. What CHIPSET?
VIA, Intel? Maker?
Asus, Acer, Intel, Gigabyte?
I would have thought this was exactly what the
test is designed to show - does the motherboard
make a difference? Does the ram speed make a
difference? If so, how much?

That way people wondering whether to go for an
expensive motherboard, with a fantastic list of
specs from the manufacturer, can take 1 minute
and 3 seconds of their time to find out how the
manufacturer's claims stack up against a real
CPU-intensive, relevant task.

Can you predict the outcome? I can't. I haven't
run a profiler over it. One thing I know though
is that the MADDR routine is highly sensitive
to adding even a single test to it - sensitive
as in you'll add 3% by doing that. My guess is
that that is memory retrieval, which would
possibly account for the fact that the figures
we are getting are not dramatically different.
From what I remember of the 386 days, you'd
upgrade processor and get a 10-fold performance
boost.

BFN. Paul.





[Non-text portions of this message have been removed]
Paul G
2010-09-06 22:59:14 UTC
Permalink
There are several variables that can change the throughput of hercules. Processor speed, number of cores, number of hardware threads, amount of memory, type of disk storage, compress on/off, etc. are among the common parts.

Hercules has lots of threads running. In addition to the one for each CPU (engine) defined, there is a logger/console thread, device threads, clock thread, network threads, control panel, HAO, disk caching, ...

This is why IBM has used a MSU or variant to determine what to charge for an OS.

A P/390E could run about 9mips, and a 4381-14 at 6.5mips, but more work gets done on the 4381 because it has higher bandwidth to devices and the P/390 almost single threaded it's IO.

For hercules, all the components of the mainframe, CPU, memory, channels, devices, clocks, etc. have to processed by the hosting OS resources. The more hardware threads, the faster you go...

On a twin xeon 5570, 2.93ghz machine with a hercules instance running a branch loop in each engine and 16 engines defined, I get about 3760 MIPS. When running a normal load, it is more like 800-1000 depending on the OS and the IO. With lots of IO activity, fewer engines is better. With smaller amounts of IO, more engines works.

While I am running the branch loop test, moving the mouse constantly will slow the machine down as much as 10%.

Lots of things can alter the performance of hercules...

Paul G.
Post by Joe Monk
Post by kerravon86
From what I remember of the 386 days, you'd
upgrade processor and get a 10-fold performance
boost.
Not anymore. You can upgrade processor, but that alone wont give you what
you are trying to achieve.
Joe
Behalf Of kerravon86
Sent: Monday, September 06, 2010 3:46 PM
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by kerravon86
Right. BUT there are so many OTHER factors, that this
test will not tell you!
That's why it is a skewed test.
For instance, lets talk motherboard design. What CHIPSET?
VIA, Intel? Maker?
Asus, Acer, Intel, Gigabyte?
I would have thought this was exactly what the
test is designed to show - does the motherboard
make a difference? Does the ram speed make a
difference? If so, how much?
That way people wondering whether to go for an
expensive motherboard, with a fantastic list of
specs from the manufacturer, can take 1 minute
and 3 seconds of their time to find out how the
manufacturer's claims stack up against a real
CPU-intensive, relevant task.
Can you predict the outcome? I can't. I haven't
run a profiler over it. One thing I know though
is that the MADDR routine is highly sensitive
to adding even a single test to it - sensitive
as in you'll add 3% by doing that. My guess is
that that is memory retrieval, which would
possibly account for the fact that the figures
we are getting are not dramatically different.
From what I remember of the 386 days, you'd
upgrade processor and get a 10-fold performance
boost.
BFN. Paul.
[Non-text portions of this message have been removed]
Mike Schwab
2010-09-06 23:44:00 UTC
Permalink
Should get about 2400MIPS on this 1U rack server.
http://www.supermicro.com/products/nfo/dunnington_intel.cfm?gclid=CJ2PtvL886MCFZZM5QodZh0k3w
4 processors * 6 cores * 2 threads * 50 MIPS per thread
--
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?
paoloG
2010-09-09 09:55:08 UTC
Permalink
Post by Mike Schwab
Should get about 2400MIPS on this 1U rack server.
http://www.supermicro.com/products/nfo/dunnington_intel.cfm?gclid=CJ2PtvL886MCFZZM5QodZh0k3w
4 processors * 6 cores * 2 threads * 50 MIPS per thread
--
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?
IMHO every core should deliver about 72 MIPS @ 2.66 Ghz (multithread is not present but shouldn't anyway give an increment in MIPS);
so with 4 procs * 6 cores * 72 MIPS *0.85 (efficiency of MP OS) = 1468 MIPS that is indeed amazing ;-)


Regards.

Paulo
Gerhard Postpischil
2010-09-06 18:50:45 UTC
Permalink
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
1:37 core i3 M330 2.13 GHz
I got 1:13 (three times) on an AMD Phenom X4, rated at 2.6 GHz



Gerhard Postpischil
Bradford, VT
Paul G
2010-09-06 20:04:38 UTC
Permalink
Post by Gerhard Postpischil
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
1:37 core i3 M330 2.13 GHz
I got 1:13 (three times) on an AMD Phenom X4, rated at 2.6 GHz
Gerhard Postpischil
Bradford, VT
Kevin E. Fischer
2010-09-06 22:34:08 UTC
Permalink
1:28 on an Intel Core 2 Quad Q6600 2.40GHz
7:26 on an Intel Atom Z520 1.33GHz Power Saver setting
5:45 on an Intel Atom Z520 1.33GHz High Performance setting

Regards
Kevin E. Fischer

-----Original Message-----
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by Gerhard Postpischil
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
1:37 core i3 M330 2.13 GHz
I got 1:13 (three times) on an AMD Phenom X4, rated at 2.6 GHz
Gerhard Postpischil
Bradford, VT
Dan Horák
2010-09-06 19:50:00 UTC
Permalink
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.
The compile is single-threaded, so it tests how
fast a single core is.
1:37 core i3 M330 2.13 GHz
It would be good if people could report speeds
on their computer.
Hercules 3.07 Fedora 12 package on x86_64
1:31 Intel Core2 Quad Q9550 @ 2.83GHz


Dan
Josef 'Jeff' Sipek
2010-09-06 20:00:21 UTC
Permalink
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.
The compile is single-threaded, so it tests how
fast a single core is.
1:37 core i3 M330 2.13 GHz
It would be good if people could report speeds
on their computer.
1:29 on my T61 with 2.5GHz Core2Duo (T9300)
--
If I have trouble installing Linux, something is wrong. Very wrong.
- Linus Torvalds
d8tahl
2010-09-06 21:08:39 UTC
Permalink
1:00 on my Intel core I7 975 3,33 ghz 12 gb ram
Post by h***@public.gmane.org
http://groups.yahoo.com/group/hercules-390/files/speed390.zip
which is a standalone package including standard
Herc 3.07 plus a single 3390 which can be IPLed
in standard S/390 mode, and do a GCC compile that
takes 2-3 minutes. ie it's a good test of CPU speed.
The compile is single-threaded, so it tests how
fast a single core is.
1:37 core i3 M330 2.13 GHz
It would be good if people could report speeds
on their computer.
Note that you can use it to test any version of
set HERCULES_RC=pdos.rc
hercules -f pdos.cnf >hercules.log
set HERCULES_RC=
I'd like to buy a new computer sometime, and I'm
interested in a laptop actually worth buying
because I can see a difference.
Thanks. Paul.
paoloG
2010-09-09 04:47:51 UTC
Permalink
Post by kerravon86
1:37 core i3 M330 2.13 GHz
It would be good if people could report speeds
on their computer.
Thanks. Paul.
My result:

58" Intel Core 2 DUO 8500 @3.16 Ghz

that means:

1) Intel Core 2 8500 @3.16Ghz is the fastest so far (that's not surprising as I chose it after testing various processors, including an Intel I7 and an Intel I5 and seeing that Core 2 8500 was faster (of course using Hercules with a single processor OS).

2) this result gives a figure of 86.2 MIPS

3) with a twin XEON 5570 @2.93 Ghz (8 cores in total) and a MP operating system you could get about 600 MIPS (IMHO).

Regards.

Paul
yvette hirth
2010-09-09 18:24:37 UTC
Permalink
my result:

HP DL380 G6 twin xeon 5540's @ 2.53Ghz, 16GB,
600 MIPs

yvette hirth
Kevin E. Fischer
2010-09-09 19:09:31 UTC
Permalink
MIN:Seconds?

Regards
Kevin E. Fischer
Mastering the art of doing very little, slowly

-----Original Message-----
From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of yvette hirth
Sent: Thursday, September 09, 2010 1:25 PM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] Re: speed test of GCCMVS on PDOS/390
my result:

HP DL380 G6 twin xeon 5540's @ 2.53Ghz, 16GB,
600 MIPs

yvette hirth


------------------------------------

</body>

<!--~-|**|PrettyHtmlStart|**|-~-->
<head>
<style type="text/css">
<!--
#ygrp-mkp {
border: 1px solid #d8d8d8;
font-family: Arial;
margin: 10px 0;
padding: 0 10px;
}

#ygrp-mkp hr {
border: 1px solid #d8d8d8;
}

#ygrp-mkp #hd {
color: #628c2a;
font-size: 85%;
font-weight: 700;
line-height: 122%;
margin: 10px 0;
}

#ygrp-mkp #ads {
margin-bottom: 10px;
}

#ygrp-mkp .ad {
padding: 0 0;
}

#ygrp-mkp .ad p {
margin: 0;
}

#ygrp-mkp .ad a {
color: #0000ff;
text-decoration: none;
}
#ygrp-sponsor #ygrp-lc {
font-family: Arial;
}

#ygrp-sponsor #ygrp-lc #hd {
margin: 10px 0px;
font-weight: 700;
font-size: 78%;
line-height: 122%;
}

#ygrp-sponsor #ygrp-lc .ad {
margin-bottom: 10px;
padding: 0 0;
}

a {
color: #1e66ae;
}

#actions {
font-family: Verdana;
font-size: 11px;
padding: 10px 0;
}

#activity {
background-color: #e0ecee;
float: left;
font-family: Verdana;
font-size: 10px;
padding: 10px;
}

#activity span {
font-weight: 700;
}

#activity span:first-child {
text-transform: uppercase;
}

#activity span a {
color: #5085b6;
text-decoration: none;
}

#activity span span {
color: #ff7900;
}

#activity span .underline {
text-decoration: underline;
}

.attach {
clear: both;
display: table;
font-family: Arial;
font-size: 12px;
padding: 10px 0;
width: 400px;
}

.attach div a {
text-decoration: none;
}

.attach img {
border: none;
padding-right: 5px;
}

.attach label {
display: block;
margin-bottom: 5px;
}

.attach label a {
text-decoration: none;
}

blockquote {
margin: 0 0 0 4px;
}

.bold {
font-family: Arial;
font-size: 13px;
font-weight: 700;
}

.bold a {
text-decoration: none;
}

dd.last p a {
font-family: Verdana;
font-weight: 700;
}

dd.last p span {
margin-right: 10px;
font-family: Verdana;
font-weight: 700;
}

dd.last p span.yshortcuts {
margin-right: 0;
}

div.attach-table div div a {
text-decoration: none;
}

div.attach-table {
width: 400px;
}

div.file-title a, div.file-title a:active, div.file-title a:hover, div.file-title a:visited {
text-decoration: none;
}

div.photo-title a, div.photo-title a:active, div.photo-title a:hover, div.photo-title a:visited {
text-decoration: none;
}

div#ygrp-mlmsg #ygrp-msg p a span.yshortcuts {
font-family: Verdana;
font-size: 10px;
font-weight: normal;
}

.green {
color: #628c2a;
}

.MsoNormal {
margin: 0 0 0 0;
}

o {
font-size: 0;
}

#photos div {
float: left;
width: 72px;
}

#photos div div {
border: 1px solid #666666;
height: 62px;
overflow: hidden;
width: 62px;
}

#photos div label {
color: #666666;
font-size: 10px;
overflow: hidden;
text-align: center;
white-space: nowrap;
width: 64px;
}

#reco-category {
font-size: 77%;
}

#reco-desc {
font-size: 77%;
}

.replbq {
margin: 4px;
}

#ygrp-actbar div a:first-child {
/* border-right: 0px solid #000;*/
margin-right: 2px;
padding-right: 5px;
}

#ygrp-mlmsg {
font-size: 13px;
font-family: Arial, helvetica,clean, sans-serif;
*font-size: small;
*font: x-small;
}

#ygrp-mlmsg table {
font-size: inherit;
font: 100%;
}

#ygrp-mlmsg select, input, textarea {
font: 99% Arial, Helvetica, clean, sans-serif;
}

#ygrp-mlmsg pre, code {
font:115% monospace;
*font-size:100%;
}

#ygrp-mlmsg * {
line-height: 1.22em;
}

#ygrp-mlmsg #logo {
padding-bottom: 10px;
}

#ygrp-mlmsg a {
color: #1E66AE;
}

#ygrp-msg p a {
font-family: Verdana;
}

#ygrp-msg p#attach-count span {
color: #1E66AE;
font-weight: 700;
}

#ygrp-reco #reco-head {
color: #ff7900;
font-weight: 700;
}

#ygrp-reco {
margin-bottom: 20px;
padding: 0px;
}

#ygrp-sponsor #ov li a {
font-size: 130%;
text-decoration: none;
}

#ygrp-sponsor #ov li {
font-size: 77%;
list-style-type: square;
padding: 6px 0;
}

#ygrp-sponsor #ov ul {
margin: 0;
padding: 0 0 0 8px;
}

#ygrp-text {
font-family: Georgia;
}

#ygrp-text p {
margin: 0 0 1em 0;
}

#ygrp-text tt {
font-size: 120%;
}

#ygrp-vital ul li:last-child {
border-right: none !important;
}
-->
</style>
</head>

<!--~-|**|PrettyHtmlEnd|**|-~-->
</html>
<!-- end group email -->
paoloG
2010-09-10 08:52:40 UTC
Permalink
Post by Kevin E. Fischer
MIN:Seconds?
Regards
Kevin E. Fischer
Mastering the art of doing very little, slowly
-----Original Message-----
Behalf Of yvette hirth
Sent: Thursday, September 09, 2010 1:25 PM
Subject: Re: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Forgive me; perhaps you don't read 58" as 58 secs (in Italy we do but I'm not sure about Anglo-Saxon countries;-)

So the official result is:

0:58 Intel Core 2 DUO 8500 @3.16 Ghz

Regards.

Paolo
Kevin E. Fischer
2010-09-10 10:43:01 UTC
Permalink
Sorry Paolo, yours I under stood, it was the " HP DL380 G6 twin xeon 5540's
@ 2.53Ghz, 16GB, 600 MIPs" that makes no sense. I should have erased your
part from my reply.

Regards
Kevin E. Fischer
Mastering the art of doing very little, slowly

-----Original Message-----
From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of paoloG
Sent: Friday, September 10, 2010 3:53 AM
To: hercules-390-***@public.gmane.org
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by Kevin E. Fischer
MIN:Seconds?
Regards
Kevin E. Fischer
Mastering the art of doing very little, slowly
-----Original Message-----
On
Post by Kevin E. Fischer
Behalf Of yvette hirth
Sent: Thursday, September 09, 2010 1:25 PM
Subject: Re: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Forgive me; perhaps you don't read 58" as 58 secs (in Italy we do but I'm
not sure about Anglo-Saxon countries;-)

So the official result is:

0:58 Intel Core 2 DUO 8500 @3.16 Ghz

Regards.

Paolo
Tony Harminc
2010-09-10 15:37:52 UTC
Permalink
Post by paoloG
Forgive me; perhaps you don't read 58" as 58 secs (in Italy we do but I'm not sure about Anglo-Saxon countries;-)
I don't know about non Anglo-Saxon countries, but here the use of '
and " for minutes and seconds is limited to the measurement of arc,
i.e. portions of the circumference of a circle, and most prominently
of the earth's circumference as used for navigation. They are never
used for the minutes and seconds of time measurement. In ordinary
notation, ' and " are used to mean feet and inches, respectively. So
there is (at least) triple overloading of the symbols.

Tony H.
paoloG
2010-09-10 18:47:02 UTC
Permalink
Post by Tony Harminc
Post by paoloG
Forgive me; perhaps you don't read 58" as 58 secs (in Italy we do but I'm not sure about Anglo-Saxon countries;-)
I don't know about non Anglo-Saxon countries, but here the use of '
and " for minutes and seconds is limited to the measurement of arc,
i.e. portions of the circumference of a circle, and most prominently
of the earth's circumference as used for navigation. They are never
used for the minutes and seconds of time measurement. In ordinary
notation, ' and " are used to mean feet and inches, respectively. So
there is (at least) triple overloading of the symbols.
Tony H.
Tony,

in Italy we use ' and " both for angles and time measurement (and of course for longitude and latitude).
In trigonometric formulas we user rads (360° = 2*greek pi rads).

Luckily we don't use miles,feet and inches ;-)
kerravon86
2010-09-11 15:08:56 UTC
Permalink
Post by paoloG
Luckily we don't use miles,feet and inches ;-)
I heard that most people in most countries spend
most of their time on websites/chat-rooms within
their own country (where they have a common
language). But one of the quirks of being in a
non-American English-speaking country is that
the American and British material is a huge
proportion of everything we see/read/hear.

And so we've been patiently (some are less patient
than others) waiting for them to convert for FRIGGIN
DECADES now. About a decade ago I saw an American
say "we're converting to metric, does anyone have
xyz conversion tool", and an Australian said "again?"
and the American said "this time it's really going
to happen". I haven't had the decadely-update since
then.

Anyway, those were quite interesting results.
The fairly small spectrum could indicate that
memory access is playing a significant role.

The top two are vastly different in cost too
(number 2 seems to be 5 times more expensive
than number 1). Pretty much shows you really
need to pick the right processor for the task
you really want. Also highlights that there's
really not that much in it for my sort of use.

Maybe for the work I do I need to buy one of
those Z9 ones I've heard other people say they
use. I regularly check Dell online, but I've
never seen them come up in the "bargain basement"
section I always check out.

I would have liked to see an i5 result too. And
the Celeron 900 would also be interesting.

Looks like my Celeron is in a fairly embarassing
position on the list. I did buy it as a disposable
computer while I was waiting for one that really
met my purpose.

I have been using this as my main guide, dividing
by the number of cores (which I have to guess a
bit).

http://www.cpubenchmark.net/cpu_list.php?t1=phenom&b1=Find+CPU

So the 8500 is about 1200, and the i7 975 is
I guess 1750, which doesn't show up as a 45%
increase in time for the 8500 by a long shot.

But with the i3 being a fairly entry level PC
I bought recently for another purpose, it
seems that we've more-or-less reached beating
dead horse territory, and I should just stick
with the cheapest deal.

BFN. Paul.



0:58 Intel Core 2 DUO 8500 @3.16 Ghz
1:00 Intel core I7 975 3.33 Ghz
1:03 xeon X5570 @2.93 Ghz
1:13 AMD Phenom X4, rated at 2.6 GHz
1:28 Intel Core 2 Quad Q6600 2.40GHz
1:29 T61 with 2.5GHz Core2Duo (T9300)
1:31 Intel Core2 Quad Q9550 @ 2.83GHz
1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz
5:45 Intel Atom Z520 1.33GHz
rhtatum
2010-09-11 16:11:17 UTC
Permalink
Interestling enough, the USA has been "offically" metric since sometime in he standard inch was defined to be exactly 2.54 centimeters and similarly for measures of area, volum and mass ... most of us aren't really aware of this. And "standard" DIPs and the like in electronic components are still dimensioned in inches. It also makes things interesting when, e.g., General Motors started converting some of their vehicles to metric. One particlarly maddening example was with the late '70s GMC Suburban - the idiots had installed the inner liners in the truck with hiddrn Allen screws so that if one wanted to remove them for the purpose of putting in soundproofing material one finally discovered that the only metric-dimensiobed fasteners on the truck were those concealed Allen screws. Ah, well.

Regards,
Ron Tatum

----- Original Message -----
From: kerravon86
To: hercules-390-***@public.gmane.org
Sent: Saturday, September 11, 2010 10:08 AM
Subject: [hercules-390] Re: speed test of GCCMVS on PDOS/390
Post by paoloG
Luckily we don't use miles,feet and inches ;-)
I heard that most people in most countries spend
most of their time on websites/chat-rooms within
their own country (where they have a common
language). But one of the quirks of being in a
non-American English-speaking country is that
the American and British material is a huge
proportion of everything we see/read/hear.

And so we've been patiently (some are less patient
than others) waiting for them to convert for FRIGGIN
DECADES now. About a decade ago I saw an American
say "we're converting to metric, does anyone have
xyz conversion tool", and an Australian said "again?"
and the American said "this time it's really going
to happen". I haven't had the decadely-update since
then.

Anyway, those were quite interesting results.
The fairly small spectrum could indicate that
memory access is playing a significant role.

The top two are vastly different in cost too
(number 2 seems to be 5 times more expensive
than number 1). Pretty much shows you really
need to pick the right processor for the task
you really want. Also highlights that there's
really not that much in it for my sort of use.

Maybe for the work I do I need to buy one of
those Z9 ones I've heard other people say they
use. I regularly check Dell online, but I've
never seen them come up in the "bargain basement"
section I always check out.

I would have liked to see an i5 result too. And
the Celeron 900 would also be interesting.

Looks like my Celeron is in a fairly embarassing
position on the list. I did buy it as a disposable
computer while I was waiting for one that really
met my purpose.

I have been using this as my main guide, dividing
by the number of cores (which I have to guess a
bit).

http://www.cpubenchmark.net/cpu_list.php?t1=phenom&b1=Find+CPU

So the 8500 is about 1200, and the i7 975 is
I guess 1750, which doesn't show up as a 45%
increase in time for the 8500 by a long shot.

But with the i3 being a fairly entry level PC
I bought recently for another purpose, it
seems that we've more-or-less reached beating
dead horse territory, and I should just stick
with the cheapest deal.

BFN. Paul.

0:58 Intel Core 2 DUO 8500 @3.16 Ghz
1:00 Intel core I7 975 3.33 Ghz
1:03 xeon X5570 @2.93 Ghz
1:13 AMD Phenom X4, rated at 2.6 GHz
1:28 Intel Core 2 Quad Q6600 2.40GHz
1:29 T61 with 2.5GHz Core2Duo (T9300)
1:31 Intel Core2 Quad Q9550 @ 2.83GHz
1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz
5:45 Intel Atom Z520 1.33GHz





[Non-text portions of this message have been removed]
Mike Schwab
2010-09-11 18:49:00 UTC
Permalink
Post by kerravon86
Post by paoloG
Luckily we don't use miles,feet and inches ;-)
I heard that most people in most countries spend
most of their time on websites/chat-rooms within
their own country (where they have a common
language). But one of the quirks of being in a
non-American English-speaking country is that
the American and British material is a huge
proportion of everything we see/read/hear.
And so we've been patiently (some are less patient
than others) waiting for them to convert for FRIGGIN
DECADES now. About a decade ago I saw an American
say "we're converting to metric, does anyone have
xyz conversion tool", and an Australian said "again?"
and the American said "this time it's really going
to happen". I haven't had the decadely-update since
then.
<deleted>

Yes, when I went to H.S. in the 1970's, there was a big effort to
teach us the metric system, since everyone else in the world uses it,
but never really got to use it since school except on food packaging.

The "If it is not American, it doesn't count" problem has bothered me
a long. I really started to notice it in the Panama Invasion against
Noriega, most media (except PBS and BBC) only reported the American
casualties and not the Panamanians. A lot of the foreign accident
news stories start out with the number of Americans killed and
injured, giving the worldwide total second or not at all.

It has continued through the Iraq and Afghanistan casualty counts, and
the ongoing ignoring of the treaty to adopt the metric system.
--
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?
yvette hirth
2010-09-11 19:47:34 UTC
Permalink
updated speed test results - last time i reported MIPs, and someone
"mentioned" that i didn't include run time. sorry!

run 1
stop 19:31:34
start 19:30:02
diff 1:32

run 2
stop 19:38:56
start 19:37:35
diff 1:21

run 3
stop 19:40:26
start 19:39:01
diff 1:25

run mean
diff 1:26

system is dual-xeon 5540's, 8 cores, 16 threads, 2.53GHz per CPU.

methinks even tho there's 16 cpu's, and there is a beneficial cache
effect for runs 2 and 3, the "unimpressive" results when compared to
other systems means that the limiting factor here is CPU clock speed.

yvette hirth
kerravon86@yahoo.com.au [hercules-390]
2015-11-14 07:12:58 UTC
Permalink
It's been a while (July 2013) since I got
any new speed tests for other processors,
and I am curious as to whether technology
has evolved in the last 2+ years.

So does anyone have a new processor
they would like to benchtest with this
program:

http://groups.yahoo.com/group/hercules-390/files/speed390.zip

Latest list shown below.

Thanks. Paul.



0:43 Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz Nvidia Geforce 8400GS
0:49 intel 2600k @ 3.4 (no OC but fast memory)
0:58 i5 650 @3.3Ghz running Win7 w 4GB
0:58 Intel Core 2 DUO 8500 @3.16 Ghz
1:00 Intel core I7 975 3.33 Ghz
1:03 Intel core I7-3632QM @2.2 Ghz
1:03 Intel core I7-2720QM @2.2 Ghz
1:03 xeon X5570 @2.93 Ghz
1:09 Intel core i5 M460 @ 2.53 GHz
1:10 Intel I5 750 @2.67 Ghz
1:13 AMD Phenom X4, rated at 2.6 GHz
1:21 dual-xeon 5540's, 8 cores, 16 threads, 2.53GHz per CPU
1:28 Intel Core 2 Quad Q6600 2.40GHz
1:29 T61 with 2.5GHz Core2Duo (T9300)
1:31 Intel Core2 Quad Q9550 @ 2.83GHz
1:37 core i3 M330 2.13 GHz
2:21 Intel Pentium Dual E2140 @ 1.60GHz
2:42 Intel Celeron M 440 @1.86 Ghz
5:45 Intel Atom Z520 1.33GHz






---In hercules-***@yahoogroups.com, <***@...> wrote :

Hello. My computer has frozen a couple of times
in the last few days, and I think it's about
time I upgraded computer to replace my Celeron M440.

I'm thinking of getting this with an Intel i5 460 M

http://configure.ap.dell.com/dellstore/config.aspx?oc=t510903au&c=au&l=en&s=dhs&cs=audhs1&model_id=inspiron-15r http://configure.ap.dell.com/dellstore/config.aspx?oc=t510903au&c=au&l=en&s=dhs&cs=audhs1&model_id=inspiron-15r

Unfortunately I don't have a result for any i5
processor, and I don't want to be disappointed
to find out that I may as well have bought an
i3, as I am basically paying extra just for the
processor.

Could someone with an i5 run the below speed test
program?

http://groups.yahoo.com/group/hercules-390/files/speed390.zip http://groups.yahoo.com/group/hercules-390/files/speed390.zip


Here's a list of what I've got results for:

0:58 Intel Core 2 DUO 8500 @3.16 Ghz
1:00 Intel core I7 975 3.33 Ghz
1:03 xeon X5570 @2.93 Ghz
1:13 AMD Phenom X4, rated at 2.6 GHz
1:21 dual-xeon 5540's, 8 cores, 16 threads, 2.53GHz per CPU
1:28 Intel Core 2 Quad Q6600 2.40GHz
1:29 T61 with 2.5GHz Core2Duo (T9300)
1:31 Intel Core2 Quad Q9550 @ 2.83GHz
1:37 core i3 M330 2.13 GHz
2:42 Intel Celeron M 440 @1.86 Ghz
5:45 Intel Atom Z520 1.33GHz

Thanks. Paul.
http://groups.yahoo.com/group/hercules-390/files/speed390.zip http://groups.yahoo.com/group/hercules-390/files/speed390.zip
Gonzalo Martin Barrio gonzalobarrio@gmail.com [hercules-390]
2015-11-17 05:06:49 UTC
Permalink
Not sure if you need just CPU specs or more, so here it goes:
MB: Gigabyte GA-Z97X-UD5H-BK
CPU: Intel Core i5 4460 3.20 GHz (stock, no OC)
RAM: 32GB (4x8GB) DDR3 G.Skill TridentX F3-2400C10D
GPU: Zotac nVidia GeForce GTX 750Ti 2GB RAM


01:50:55 01:50:55 * MSG FROM HERCULES: COPY PDPTOP
...
01:51:38 01:51:38 * MSG FROM HERCULES:
END
...
01:51:38 01:51:38 * MSG FROM HERCULES: return from PCOMM is
0
----------------------------------
0:43
Post by ***@yahoo.com.au [hercules-390]
So does anyone have a new processor
they would like to benchtest with this
Forgot to mention - this is on Windows.
BFN. Paul.
kerravon86@yahoo.com.au [hercules-390]
2015-11-17 05:39:18 UTC
Permalink
Post by Gonzalo Martin Barrio ***@gmail.com [hercules-390]
Intel Core i5 4460 3.20 GHz (stock, no OC)
Thanks Gonzalo, I have added that
to my list.

BFN. Paul.




---In hercules-***@yahoogroups.com, <***@...> wrote :

Not sure if you need just CPU specs or more, so here it goes:
MB: Gigabyte GA-Z97X-UD5H-BK
CPU: Intel Core i5 4460 3.20 GHz (stock, no OC)

RAM: 32GB (4x8GB) DDR3 G.Skill TridentX F3-2400C10D

GPU: Zotac nVidia GeForce GTX 750Ti 2GB RAM


01:50:55 01:50:55 * MSG FROM HERCULES: COPY PDPTOP
...
01:51:38 01:51:38 * MSG FROM HERCULES: END
...
01:51:38 01:51:38 * MSG FROM HERCULES: return from PCOMM is 0
----------------------------------
0:43
Post by Gonzalo Martin Barrio ***@gmail.com [hercules-390]
So does anyone have a new processor
they would like to benchtest with this
Forgot to mention - this is on Windows.

BFN. Paul.
opplr@hotmail.com [hercules-390]
2015-12-06 20:25:13 UTC
Permalink
Paul wrote:

"It's been a while (July 2013) since I got
any new speed tests for other processors,
and I am curious as to whether technology
has evolved in the last 2+ years.

So does anyone have a new processor
they would like to benchtest with this
program:"

-----------------------------------------------------------

Are you also keeping up with which version of Hercules was used for the test ?

IE 3.07, 3.08, 3.09, 3.10, 3.11 & now 3.12 ?

Phil
opplr@hotmail.com [hercules-390]
2015-12-07 01:08:48 UTC
Permalink
Earlier I wrote:

"Are you also keeping up with which version of Hercules was used for the test ?

IE 3.07, 3.08, 3.09, 3.10, 3.11 & now 3.12 ?"

---------------------------------------------------------------------------------------------------------------------

It has been too long. I see now the speed test is suppose to be self-contained. Had to copy MSVCxx80.dll to Win2K to test with.

Not slowest laptop I have but many benchmarks have been run on it.

16:34 IBM ThinkPad 390e Pentium II 300 MHz Win2K

16:18 IBM ThinkPad 390e Pentium II 300 MHz Win2k Hercules 3.11

Other old laptops:

12:35 NoName Notebook Celeron 400 MHz WIN2k

12:35 NoName Notebook Celeron 400 MHz WIN2k Hercules 3.07

11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2

Phil

ps - yes those are minutes and seconds Hercules 3.02 can run the test but puts out an extra message that MSG (or similar) isn't a valid command so timing would have been off due to that.
kerravon86@yahoo.com.au [hercules-390]
2015-12-07 01:33:41 UTC
Permalink
Hi Phil.

Thanks for the new data.

BFN. Paul.






0:43 Intel Core i5 4460 3.20 GHz (stock, no OC)
0:43 Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz Nvidia Geforce 8400GS
0:49 intel 2600k @ 3.4 (no OC but fast memory)
0:58 i5 650 @3.3Ghz running Win7 w 4GB
0:58 Intel Core 2 DUO 8500 @3.16 Ghz
1:00 Intel core I7 975 3.33 Ghz
1:03 Intel core I7-3632QM @2.2 Ghz
1:03 Intel core I7-2720QM @2.2 Ghz
1:03 xeon X5570 @2.93 Ghz
1:09 Intel core i5 M460 @ 2.53 GHz
1:10 Intel I5 750 @2.67 Ghz
1:13 AMD Phenom X4, rated at 2.6 GHz
1:21 dual-xeon 5540's, 8 cores, 16 threads, 2.53GHz per CPU
1:28 Intel Core 2 Quad Q6600 2.40GHz
1:29 T61 with 2.5GHz Core2Duo (T9300)
1:31 Intel Core2 Quad Q9550 @ 2.83GHz
1:37 core i3 M330 2.13 GHz
2:21 Intel Pentium Dual E2140 @ 1.60GHz
2:42 Intel Celeron M 440 @1.86 Ghz
5:45 Intel Atom Z520 1.33GHz
11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2
12:35 NoName Notebook Celeron 400 MHz WIN2k
16:34 IBM ThinkPad 390e Pentium II 300 MHz Win2K




---In hercules-***@yahoogroups.com, <***@...> wrote :

Earlier I wrote:

"Are you also keeping up with which version of Hercules was used for the test ?

IE 3.07, 3.08, 3.09, 3.10, 3.11 & now 3.12 ?"

---------------------------------------------------------------------------------------------------------------------

It has been too long. I see now the speed test is suppose to be self-contained. Had to copy MSVCxx80.dll to Win2K to test with.

Not slowest laptop I have but many benchmarks have been run on it.

16:34 IBM ThinkPad 390e Pentium II 300 MHz Win2K

16:18 IBM ThinkPad 390e Pentium II 300 MHz Win2k Hercules 3.11

Other old laptops:

12:35 NoName Notebook Celeron 400 MHz WIN2k

12:35 NoName Notebook Celeron 400 MHz WIN2k Hercules 3.07

11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2

Phil

ps - yes those are minutes and seconds Hercules 3.02 can run the test but puts out an extra message that MSG (or similar) isn't a valid command so timing would have been off due to that.
opplr@hotmail.com [hercules-390]
2015-12-07 06:13:42 UTC
Permalink
Earlier I wrote something incorrectly:

"11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2"

That should be a 'Dell Latitude LS'

And another entrant in the slowest of the slow:

33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat hercules port 6 watts

This is a little off the beaten path as I picked it up for about $18 .

It has only 256 Meg, no hard drive, 4 USB 2.0 ports. Juergen cross-compiled the arm-softfloat hercules executables from a much later distribution so bits and pieces of Debian Squeeze 2.6.32-5 had to be upgraded so hercules would run.

Boot Debian from a USB stick and run the speed test from a USB IDE hard drive while connected as a console device via Putty over WiFi.

Plenty of room for slowness to come into the picture.

Phil
kerravon86@yahoo.com.au [hercules-390]
2015-12-07 06:22:56 UTC
Permalink
I've updated the first.

The second sounds like it's not using the
executable provided with speed390. I
am really looking for processors running
that exact executable, because I want
to see how fast I can run that executable,
so I know which processor to buy. Or
what type of memory, as I have a
suspicion that faster memory within
the same processor will make a
significant difference.

BFN. Paul.





---In hercules-***@yahoogroups.com, <***@...> wrote :

Earlier I wrote something incorrectly:

"11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2"

That should be a 'Dell Latitude LS'

And another entrant in the slowest of the slow:

33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat hercules port 6 watts

This is a little off the beaten path as I picked it up for about $18 .

It has only 256 Meg, no hard drive, 4 USB 2.0 ports. Juergen cross-compiled the arm-softfloat hercules executables from a much later distribution so bits and pieces of Debian Squeeze 2.6.32-5 had to be upgraded so hercules would run.

Boot Debian from a USB stick and run the speed test from a USB IDE hard drive while connected as a console device via Putty over WiFi.

Plenty of room for slowness to come into the picture.

Phil
opplr@hotmail.com [hercules-390]
2015-12-07 13:14:00 UTC
Permalink
Paul wrote:

"The second sounds like it's not using the
executable provided with speed390. I
am really looking for processors running
that exact executable, because I want
to see how fast I can run that executable,
so I know which processor to buy. Or
what type of memory, as I have a
suspicion that faster memory within
the same processor will make a
significant difference."

---------------------------------------------------------

The PogoPlug test does not use the executable included in speed390. It is gcc compiled on a fairly recent Ubuntu distro IIRC.

Hmmm,

Larger cache sizes on the processor core may make the biggest difference.

Faster memory may be a factor but hard drive I/O seems to be the biggest limiting factor for speed.

Don't have a SSD, but pehaps some one can run the test on the same machine with one vs a hard drive and see if there is much of a difference or not.

Phil
kerravon86@yahoo.com.au [hercules-390]
2015-12-08 03:01:18 UTC
Permalink
Post by ***@hotmail.com [hercules-390]
Faster memory may be a factor but
hard drive I/O seems to be the
biggest limiting factor for speed.
I'm not sure what environment you are
talking about, but on normal laptops,
hard drive I/O is irrelevant. It is totally
CPU-bound (or memory-bound).

You can see this for yourself if you
run with "-O0" instead of "-Os". The
unoptimized compile (which reads
the exact same input and produces
even more output) flies through
the compile. Optimization brings
it to a crawl.

BFN. Paul.





---In hercules-***@yahoogroups.com, <***@...> wrote :

Paul wrote:

"The second sounds like it's not using the
executable provided with speed390. I
am really looking for processors running
that exact executable, because I want
to see how fast I can run that executable,
so I know which processor to buy. Or
what type of memory, as I have a
suspicion that faster memory within
the same processor will make a
significant difference."

---------------------------------------------------------

The PogoPlug test does not use the executable included in speed390. It is gcc compiled on a fairly recent Ubuntu distro IIRC.

Hmmm,

Larger cache sizes on the processor core may make the biggest difference.

Faster memory may be a factor but hard drive I/O seems to be the biggest limiting factor for speed.

Don't have a SSD, but pehaps some one can run the test on the same machine with one vs a hard drive and see if there is much of a difference or not.

Phil
opplr@hotmail.com [hercules-390]
2015-12-08 05:41:39 UTC
Permalink
Paul wrote:



"I'm not sure what environment you are
talking about, but on normal laptops,
hard drive I/O is irrelevant. It is totally
CPU-bound (or memory-bound).

You can see this for yourself if you
run with "-O0" instead of "-Os"."


-----------------------------------------------------------------


Think I got fixated on the PogoPlug performance being so dismal. Yes, GCCMVS is usually CPU bound. Something else seems to be happening with 'abnormal' environments like PogoPlug or perhaps the Raspberry Pi (?).


Anyway a quick test of IMON's instruction rate panel on both the 1.2 GHz ARM and 300 MHz Pentium II doesn't seem to explain how the 300 MHz speed390 result was 2 times as fast as the 4 times faster ARM.


33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat hercules port 6 watts

16:34 IBM ThinkPad 390e Pentium II 300 MHz Win2K


IMON rate panel:


300 MHz Pentium II - IMON Test - 1.2 GHz ARM

8,619,872 TEST 1-RR(F): 9,917,879
5,752,945 TEST 2-RR(S): 4,996,502
4,364,601 TEST 3-RX(F): 2,817,599
1,404,644 TEST 4-RX(S): 1,161,030
710,498 TEST 5-SS(F): 1,083,241
203,767 TEST 6-SS(S): 429,335
2,302,471 TEST 7-FP(E): 2,285,682
428,821 TEST 8-FP(D): 813,027


I have yet to figure out how to keep the formating on the new yahoo.


Phil


ps - Win2K used Hercules 3.07 ARM Linux used TK4-binaries which are not the same as Hercules 3.07 so some variation in instruction speed can be expected.
Vince Coen vbcoen@gmail.com [hercules-390]
2015-12-07 13:47:06 UTC
Permalink
Where is this speed test and on what platform & software combination
does it run on?
Post by ***@yahoo.com.au [hercules-390]
I've updated the first.
The second sounds like it's not using the
executable provided with speed390. I
am really looking for processors running
that exact executable, because I want
to see how fast I can run that executable,
so I know which processor to buy. Or
what type of memory, as I have a
suspicion that faster memory within
the same processor will make a
significant difference.
BFN. Paul.
"11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2"
That should be a 'Dell Latitude LS'
33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat hercules port 6 watts
This is a little off the beaten path as I picked it up for about $18 .
It has only 256 Meg, no hard drive, 4 USB 2.0 ports. Juergen
cross-compiled the arm-softfloat hercules executables from a much
later distribution so bits and pieces of Debian Squeeze 2.6.32-5 had
to be upgraded so hercules would run.
Boot Debian from a USB stick and run the speed test from a USB IDE
hard drive while connected as a console device via Putty over WiFi.
Plenty of room for slowness to come into the picture.
Phil
opplr@hotmail.com [hercules-390]
2015-12-07 15:02:22 UTC
Permalink
Vince wrote:

"Where is this speed test and on what platform & software combination
does it run on?"

-------------------------------------------------------------------------------------------------------

It's located in the files section of this group

speed390.zip

It is self contained and set up for Windows systems. Hercules does not have to be installed. Just unzip to a directory, change to directory and 'doit' bat file should start it off.

It prefers WinXP or later since all the Visual C runtimes are present, but can be run on Win2K with addition of one of the MSVCxx80.dll files.

It can also be run on any platform where Hercules is running. Just adjust the doit.bat file to fit Linux or Mac, etc...

Paul generally has opted for laptops which have a little different performance than desktops.

Earlier I mentioned cache perhaps being the #1 impactor in performance. That is if I/O isn't being a PITA.

Use google translate to check out Intel & AMD cache differences on various CPU for gaming performance at:

www.iyd.kr/695 http://www.iyd.kr/695

I have had one of those G3258 chips for over a year now and haven't built the system yet.

Phil
Vince Coen vbcoen@gmail.com [hercules-390]
2015-12-07 16:49:19 UTC
Permalink
Didn't do to well here and this is from the log file

-----------
16:37:14 HHC01413I Hercules version 4.00
16:37:14 HHC01414I (c) Copyright 1999-2015 by Roger Bowler, Jan Jaeger,
and others
16:37:14 HHC01415I Build date: Nov 19 2015 at 23:05:48
16:37:14 HHC01417I Built with: GCC 4.9.2
16:37:14 HHC01417I Build type: GNU/Linux x86_64 host architecture build
16:37:14 HHC01417I Modes: S/370 ESA/390 z/Arch
16:37:14 HHC01417I Max CPU Engines: 8
16:37:14 HHC01417I Using setresuid() for setting privileges
16:37:14 HHC01417I Using POSIX threads Threading Model
16:37:14 HHC01417I Using Error-Checking Mutex Locking Model
16:37:14 HHC01417I With Syncio support
16:37:14 HHC01417I With Shared Devices support
16:37:14 HHC01417I With Dynamic loading support
16:37:14 HHC01417I Using shared libraries
16:37:14 HHC01417I With External GUI support
16:37:14 HHC01417I With IPV6 support
16:37:14 HHC01417I With HTTP Server support
16:37:14 HHC01417I With sqrtl support
16:37:14 HHC01417I With SIGABEND handler
16:37:14 HHC01417I With CCKD BZIP2 support
16:37:14 HHC01417I With HET BZIP2 support
16:37:14 HHC01417I With ZLIB support
16:37:14 HHC01417I With Regular Expressions support
16:37:14 HHC01417I With Object REXX support
16:37:14 HHC01417I With Regina REXX support
16:37:14 HHC01417I With Automatic Operator support
16:37:14 HHC01417I Without National Language Support
16:37:14 HHC01417I Machine dependent assists: cmpxchg1 cmpxchg4 cmpxchg8
hatomics=gccIntrinsics
16:37:14 HHC01417I Running on: Applewood (Linux-4.1.13-server-2.mga5
x86_64) MP=8
16:37:14 HHC01508I HDL: loadable module directory is /usr/local/lib/hercules
16:37:14 HHC00150I Crypto module loaded (c) Copyright 2003-2015 by
Bernard van der Helm
16:37:14 HHC00151I Activated facility: Message Security Assist
16:37:14 HHC00151I Activated facility: Message Security Assist Extension
1, 2, 3 and 4
16:37:14 HHC00100I Thread id 09d48700, prio 2147483647, name Processor
CP00 started
16:37:14 HHC00100I Thread id 09c47700, prio 2147483647, name Timer started
16:37:14 HHC00811I Processor CP00: architecture mode z/Arch
16:37:14 HHC17003I MAIN storage is 512M (mainsize); storage is not
locked
16:37:14 HHC00811I Processor CP00: architecture mode ESA/390
16:37:14 HHC02204I archmode set to ESA/390
16:37:14 HHC02204I cpuserial set to 000611
16:37:14 HHC02204I cpumodel set to 4381
16:37:14 HHC02204I numcpu set to 1
16:37:14 HHC00414I 0:01B9 CKD file pdos00.1b9: cyls 1113 heads 15 tracks
16695 trklen 56832
16:37:14 HHC00100I Thread id 0d316700, prio 2147483647, name Control
panel started
16:37:14 HHC02260I Script 1: begin processing file pdos.rc
16:37:14 HHC01603I ipl 1b9
16:37:15 HHC02264I Script 1: file pdos.rc processing ended
16:37:15 HHC00100I Thread id 09e4a700, prio 2147483647, name Read-ahead
thread-1 started
16:37:15 HHC00100I Thread id 0993c700, prio 2147483647, name Read-ahead
thread-2 started
16:37:15 16:37:15 * MSG FROM HERCULES: PDOS should reside on cylinder
1, head 0 of IPL device
16:37:15 16:37:15 * MSG FROM HERCULES: IPL device is
10000
16:37:15 16:37:15 * MSG FROM HERCULES: Welcome to
PDOS!!!
16:37:15 16:37:15 * MSG FROM HERCULES: CR0 is
00B00000
16:37:15 16:37:15 * MSG FROM HERCULES: PDOS structure is 360544
bytes
16:37:15 16:37:15 * MSG FROM HERCULES: aspace padding is 7808
bytes
16:37:15 16:37:15 * MSG FROM HERCULES: IPL device is
10000
16:37:15 16:37:15 * MSG FROM HERCULES: aspace 0, seg 00319000, cr13
00000000
16:37:15 16:37:15 * MSG FROM HERCULES: aspace 1, seg 0032F000, cr13
00000000
16:37:15 16:37:15 * MSG FROM HERCULES: aspace 2, seg 00345000, cr13
00000000
16:37:15 16:37:15 * MSG FROM HERCULES: aspace 3, seg 0035B000, cr13
00000000
16:37:15 16:37:15 * MSG FROM HERCULES: PCOMM should reside on cylinder
2, head 0 of IPL device
16:37:15 HHC00822I Processor CP00: machine check due to host error:
Segmentation fault
16:37:15 HHC02324I PSW=000C0000802159BA INST=5820C28E L
2,654(0,12) load
16:37:15 HHC02326I R:00215A8C:K:06=00008010 FFFFFFFF 00008014 00215744
................
16:37:15 HHC02269I GR00=00000000 GR01=00100840 GR02=00504814 GR03=00004814
16:37:15 HHC02269I GR04=00100868 GR05=00004814 GR06=00008088 GR07=00008090
16:37:15 HHC02269I GR08=00000000 GR09=00000460 GR10=00215B10 GR11=001007E0
16:37:15 HHC02269I GR12=802157FE GR13=001007E8 GR14=802158AA GR15=00004814
16:37:46 HHC01603I exit
16:37:46 HHC01420I Begin Hercules shutdown
16:37:46 HHC01423I Calling termination routines
16:37:46 HHC00101I Thread id 09c47700, prio 2147483647, name Timer ended
16:37:46 HHC02272I Highest observed MIPS and IO/s rates
16:37:46 HHC02272I from Mon Dec 7 00:00:00 2015
16:37:46 HHC02272I to Mon Dec 7 16:37:46 2015
16:37:46 HHC02272I MIPS: 2.338800 IO/s: 72
16:37:46 HHC00101I Thread id 09d48700, prio 2147483647, name Processor
CP00 ended
16:37:46 HHC00333I 0:01B9 size free nbr st reads writes
l2reads hits switches
16:37:46 HHC00334I
0:01B9 readaheads misses
16:37:46 HHC00335I 0:01B9
--------------------------------------------------------------------
16:37:46 HHC00336I 0:01B9 [*] 0001314166 000 % 0002 0000007 0000000
0000001 0000022 0000026
16:37:46 HHC00337I
0:01B9 0000024 0000000
16:37:46 HHC00338I 0:01B9 pdos00.1b9
16:37:46 HHC00339I 0:01B9 [0] 0001314166 000 % 0002 rw 0000007 0000000
0000001
16:37:46 HHC00101I Thread id 0993c700, prio 2147483647, name Read-ahead
thread-2 ended
16:37:46 HHC00101I Thread id 09e4a700, prio 2147483647, name Read-ahead
thread-1 ended
16:37:46 HHC01427I Main storage released
16:37:46 HHC01427I Expanded storage released
16:37:46 HHC01422I Configuration released
-----------

Run manually using :

hercules -f pdos.cnf -r pdos.rc >hercules.log


Vince
Post by ***@hotmail.com [hercules-390]
"Where is this speed test and on what platform & software combination
does it run on?"
-------------------------------------------------------------------------------------------------------
It's located in the files section of this group
speed390.zip
It is self contained and set up for Windows systems. Hercules does
not have to be installed. Just unzip to a directory, change to
directory and 'doit' bat file should start it off.
It prefers WinXP or later since all the Visual C runtimes are present,
but can be run on Win2K with addition of one of the MSVCxx80.dll files.
It can also be run on any platform where Hercules is running. Just
adjust the doit.bat file to fit Linux or Mac, etc...
Paul generally has opted for laptops which have a little different
performance than desktops.
Earlier I mentioned cache perhaps being the #1 impactor in
performance. That is if I/O isn't being a PITA.
Use google translate to check out Intel & AMD cache differences on
www.iyd.kr/695 <http://www.iyd.kr/695>
I have had one of those G3258 chips for over a year now and haven't built the system yet.
Phil
------------------------------------------------------------------------
opplr@hotmail.com [hercules-390]
2015-12-07 18:04:56 UTC
Permalink
Vince wrote about trouble:

"16:37:15 16:37:15 * MSG FROM HERCULES: PCOMM should reside on cylinder
2, head 0 of IPL device
16:37:15 HHC00822I Processor CP00: machine check due to host error:
Segmentation fault
16:37:15 HHC02324I PSW=000C0000802159BA INST=5820C28E L
2,654(0,12) load
16:37:15 HHC02326I R:00215A8C:K:06=00008010 FFFFFFFF 00008014 00215744"

-------------------------------------------------------------------------------------------------------------

Since it does actually execute some instructions from the pdos DASD maybe speed390 isn't ready for :

Build type: GNU/Linux x86_64 host architecture build

I have been running with x86 32 bit binaries ( other than the ARM test ).

Phil
opplr@hotmail.com [hercules-390]
2015-12-07 19:45:35 UTC
Permalink
Earlier I wrote:

"maybe speed390 isn't ready for :

Build type: GNU/Linux x86_64 host architecture build"

-------------------------------------------------------------------------------

Something else is going on.

There are 3 mythbuntu boxes in the house here. Since apt-get hercules on each pulls in 3.07-x-x I tried speed390 on all 3.

All desktops, 2 i686 and one x86_64.

1.13 AMD Phenom II X3 720 2.8 Ghz 3 cores i686

2.13 Intel Core2 Duo E4300 1.8 GHz 2 cores 2 threads x86_64

4:05 Intel Atom 330 1.6 GHz 2 cores 4 threads i686 8 watts

Believe only change was in pdos.cnf PANRATE MEDIUM changed to PANRATE 500 .

So x86_64 does run the speed390 test with Hercules 3.07 or similar ( TK4- hercules binaries ). Don't know which 4.0 snapshot you are trying to run with.

Phil

ps - all these are Linux so outside of Paul's particular interest ( I think ). The thing about speed390 is the amount of I/O it actually engages in. The PogoPlug test shows this to the extreme (TK4- IPLs fairly quickly). Console message has to go to USB IDE drive for the log file and same messages go to putty instance via WiFi connection.

If more time is available maybe I'll connect a old drive to these 3 and load up windows for results.

pps - Paul how about adding 'find "return from PCOMM" hercules.log' and 'find "COPY PDPTOP" hercules.log to the bat file for auto display of the times you are looking for.
kerravon86@yahoo.com.au [hercules-390]
2015-12-08 02:56:37 UTC
Permalink
Post by ***@hotmail.com [hercules-390]
pps - Paul how about adding
'find "return from PCOMM" hercules.log'
and 'find "COPY PDPTOP" hercules.log
to the bat file for auto display of the
times you are looking for.
Thanks for the suggestion Phil. I
was not aware of the "find"
command in Windows, so I
have just tried it out and it seems
to be fine.

I'll make that change if I reissue
speed390 for some reason.

BFN. Paul.




---In hercules-***@yahoogroups.com, <***@...> wrote :

Earlier I wrote:

"maybe speed390 isn't ready for :

Build type: GNU/Linux x86_64 host architecture build"

-------------------------------------------------------------------------------

Something else is going on.

There are 3 mythbuntu boxes in the house here. Since apt-get hercules on each pulls in 3.07-x-x I tried speed390 on all 3.

All desktops, 2 i686 and one x86_64.

1.13 AMD Phenom II X3 720 2.8 Ghz 3 cores i686

2.13 Intel Core2 Duo E4300 1.8 GHz 2 cores 2 threads x86_64

4:05 Intel Atom 330 1.6 GHz 2 cores 4 threads i686 8 watts

Believe only change was in pdos.cnf PANRATE MEDIUM changed to PANRATE 500 .

So x86_64 does run the speed390 test with Hercules 3.07 or similar ( TK4- hercules binaries ). Don't know which 4.0 snapshot you are trying to run with.

Phil

ps - all these are Linux so outside of Paul's particular interest ( I think ). The thing about speed390 is the amount of I/O it actually engages in. The PogoPlug test shows this to the extreme (TK4- IPLs fairly quickly). Console message has to go to USB IDE drive for the log file and same messages go to putty instance via WiFi connection.

If more time is available maybe I'll connect a old drive to these 3 and load up windows for results.

pps - Paul how about adding 'find "return from PCOMM" hercules.log' and 'find "COPY PDPTOP" hercules.log to the bat file for auto display of the times you are looking for.
opplr@hotmail.com [hercules-390]
2015-12-07 16:57:12 UTC
Permalink
Earlier I wrote:

"It prefers WinXP or later"

-----------------------------------

Correction it prefers WinXPsp2 or later. WinXPsp1 had to have the Visual C runtime 2005sp1 installed before it would work ( and possibly the 2008sp1 runtimes also as I did them before the 2005 one).

More entries:

2:00 HP Pavilion ze4805us AMD Athlon XP-M 2800+ (2.133GHz) WinXPsp1
2:12 IBM ThinkPad z60m Intel Pentium M 740 1.7 GHz WinXPsp2
2:30 Gateway 4520 Intel Pentium M 705 1.5 GHzWinXPsp3

Phil
Robert Prins robert.ah.prins@gmail.com [hercules-390]
2015-12-07 17:59:02 UTC
Permalink
3.5GHz AMD 8-core FX8150 with 16GB:


18:50:58 Hercules Version 3.07
18:50:58 (c)Copyright 1999-2010 by Roger Bowler, Jan Jaeger, and others
18:50:58 Built on Mar 23 2010 at 01:39:37
18:50:58 Build information:
18:50:58 Windows (MSVC) build for i386
18:50:58 Modes: S/370 ESA/390 z/Arch
18:50:58 Max CPU Engines: 8
18:50:58 Using fthreads instead of pthreads
18:50:58 Dynamic loading support
18:50:58 Using shared libraries
18:50:58 HTTP Server support
18:50:58 No SIGABEND handler
18:50:58 Regular Expressions support
18:50:58 Automatic Operator support
18:50:58 Machine dependent assists: cmpxchg1 cmpxchg4 cmpxchg8 fetch_dw
store_dw
18:50:58 Running on MONSTER Windows_NT-6.1 i686 MP=8
18:50:58 HHCHD018I Loadable module directory is hercules
18:50:58 HHCCF065I Hercules: tid=00002254, pid=816, pgid=816, priority=0
18:50:58 HHCDA020I pdos00.1b9 cyls=1113 heads=15 tracks=16695 trklen=56832
18:50:58 HHCCP002I CPU0000 thread started: tid=00001EF8, pid=816,
priority=15
18:50:58 HHCCP003I CPU0000 architecture mode ESA/390
18:50:58 HHCTT002I Timer thread started: tid=000019A8, pid=816, priority=0
18:50:58 HHCPN001I Control panel thread started: tid=00002254, pid=816
18:50:58 HHCPN008I Script file processing started using file "pdos.rc"
18:50:58 HHCAO001I Hercules Automatic Operator thread started;
18:50:58 tid=000019DC, pri=0, pid=816
18:50:58 ipl 1b9
18:50:58 HHCPN013I EOF reached on SCRIPT file. Processing complete.
18:50:58 18:50:58 * MSG FROM HERCULES: PDOS should reside on cylinder 1,
head 0 of IPL device
18:50:58 18:50:58 * MSG FROM HERCULES: IPL device is 10000
18:50:58 18:50:58 * MSG FROM HERCULES: Welcome to PDOS!!!
18:50:58 18:50:58 * MSG FROM HERCULES: CR0 is 00B00000
18:50:58 18:50:58 * MSG FROM HERCULES: PDOS structure is 360544 bytes
18:50:58 18:50:58 * MSG FROM HERCULES: aspace padding is 7808 bytes
18:50:58 18:50:58 * MSG FROM HERCULES: IPL device is 10000
18:50:58 18:50:58 * MSG FROM HERCULES: aspace 0, seg 00319000, cr13
00000000
18:50:58 18:50:58 * MSG FROM HERCULES: aspace 1, seg 0032F000, cr13
00000000
18:50:58 18:50:58 * MSG FROM HERCULES: aspace 2, seg 00345000, cr13
00000000
18:50:58 18:50:58 * MSG FROM HERCULES: aspace 3, seg 0035B000, cr13
00000000
18:50:58 18:50:58 * MSG FROM HERCULES: PCOMM should reside on cylinder 2,
head 0 of IPL device
18:50:58 18:50:58 * MSG FROM HERCULES: probably last track on cylinder
18:50:58 18:50:58 * MSG FROM HERCULES: probably last track on cylinder
18:50:58 18:50:58 * MSG FROM HERCULES: probably last track on cylinder
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 24
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 64
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 27
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 22
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 24
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 64
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 27
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 22
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 24
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 64
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 27
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 22
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 120
18:50:58 18:50:58 * MSG FROM HERCULES: SVC code is 10
18:50:58 18:50:58 * MSG FROM HERCULES: COPY PDPTOP
.
.
.
18:51:57 18:51:57 * MSG FROM HERCULES: END
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 20
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 20
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 20
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 10
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 120
18:51:57 18:51:57 * MSG FROM HERCULES: SVC code is 3
18:51:57 18:51:57 * MSG FROM HERCULES: return from PCOMM is 0
18:53:05 quit
18:53:05 HHCIN900I Begin Hercules shutdown
18:53:05 HHCIN901I Releasing configuration
18:53:05 HHCCP008I CPU0000 thread ended: tid=00001EF8, pid=816
18:53:05 HHCAO002I Hercules Automatic Operator thread ended
18:53:05 HHCTT003I Timer thread ended
18:53:05 HHCCD210I size free nbr st reads writes l2reads
hits switches
18:53:05 HHCCD212I
--------------------------------------------------------------------
18:53:05 HHCCD213I [*] 1314166 0% 2 65 0
1 0 84
18:53:05 HHCCD215I pdos00.1b9
18:53:05 HHCCD216I [0] 1314166 0% 2 rw 65 0 1
18:53:05 HHCCF047I Subchannel 0:0000 detached
18:53:05 HHCCF047I Subchannel 0:0001 detached
18:53:05 HHCIN902I Configuration release complete
18:53:05 HHCIN903I Calling termination routines
18:53:05 HHCHD900I Begin shutdown sequence
18:53:05 HHCHD901I Calling panel_cleanup
18:53:05 HHCHD902I panel_cleanup complete
18:53:05 HHCLG014I logger thread terminating

59 seconds, with z/OS 1.10 idle in another process.

Notebook (2.5GHz Intel i7 MQ4710 with 24Gb) will come later.

Can try them also on V3.11, still a bit reluctant to go to V3.12, although
I've not heard anything bad about it...

Robert
Post by ***@hotmail.com [hercules-390]
"It prefers WinXP or later"
-----------------------------------
Correction it prefers WinXPsp2 or later. WinXPsp1 had to have the Visual
C runtime 2005sp1 installed before it would work ( and possibly the 2008sp1
runtimes also as I did them before the 2005 one).
2:00 HP Pavilion ze4805us AMD Athlon XP-M 2800+ (2.133GHz) WinXPsp1
2:12 IBM ThinkPad z60m Intel Pentium M 740 1.7 GHz WinXPsp2
2:30 Gateway 4520 Intel Pentium M 705 1.5 GHzWinXPsp3
Phil
--
Robert AH Prins
***@gmail.com
opplr@hotmail.com [hercules-390]
2015-12-07 20:32:28 UTC
Permalink
An additional Linux entry: ( a no longer used for mythtv Mythbuntu box ).

0.56 AMD FX-8320 8 Cores 3.5 GHz x86_64 ( full turbo mode 3.7 GHz; half turbo mode 4.0 GHz )

Whatever half and full turbo modes are.

Phil
Tony Harminc tharminc@gmail.com [hercules-390]
2015-12-07 20:49:11 UTC
Permalink
Win 7 64-bit with various minor background desktop stuff going on, so Task
Manager is typically flickering around 5% CPU..

Rating: 5.9 Windows Experience Index
Processor: Intel(R) Core(TM) i7-2600K CPU @ 3.4GHz 3.70 GHz
Installed memory(RAM): 16.0 GB
System type: 64-bit Operating System
15:34:25 15:34:25 * MSG FROM HERCULES: COPY PDPTOP
15:35:18 15:35:18 * MSG FROM HERCULES: END

Which is 53 seconds, if my time arithmetic can be trusted.

Tony H.
Gregg Levine gregg.drwho8@gmail.com [hercules-390]
2015-12-07 07:08:36 UTC
Permalink
Hello!
Phil, when you get some free time (whatever that is!) would you please
document how you accomplished this one:
33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat
hercules port 6 watts

Your saying:
"It has only 256 Meg, no hard drive, 4 USB 2.0 ports. Juergen cross-compiled
the arm-softfloat hercules executables from a much later distribution so
bits and pieces of Debian Squeeze 2.6.32-5 had to be upgraded so hercules
would run.

Boot Debian from a USB stick and run the speed test from a USB IDE hard
drive while connected as a console device via Putty over WiFi.

Plenty of room for slowness to come into the picture."

I have here three of the things, currently one who's a close second to
that one is not connected, and one who's a close third is also, One is
in pink, that's the original design, and the other in black calls
itself a business oriented PogoPlug, but I'm not so sure of that. I
originally bought the first to do that, then switched to doing what it
does now, wear drives, and bought the others for the intended purpose.
And as it happens the fact that it runs Linux inside bothers the help
desk, they would want us to not know that......

Please document it in a new thread. I believe some of us would be
interested. My target would have been the guy in pink, if they weren't
so resistant to the idea.
-----
Gregg C Levine ***@gmail.com
"This signature fought the Time Wars, time and again."
Post by ***@hotmail.com [hercules-390]
"11:49 Dell Latitude ES Pentium III 400 MHz WINXPsp2"
That should be a 'Dell Latitude LS'
33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat hercules port 6 watts
This is a little off the beaten path as I picked it up for about $18 .
It has only 256 Meg, no hard drive, 4 USB 2.0 ports. Juergen cross-compiled
the arm-softfloat hercules executables from a much later distribution so
bits and pieces of Debian Squeeze 2.6.32-5 had to be upgraded so hercules
would run.
Boot Debian from a USB stick and run the speed test from a USB IDE hard
drive while connected as a console device via Putty over WiFi.
Plenty of room for slowness to come into the picture.
Phil
________________________________
________________________________
opplr@hotmail.com [hercules-390]
2015-12-07 13:23:33 UTC
Permalink
Greg wrote;

"Phil, when you get some free time (whatever that is!) would you please
document how you accomplished this one:
33:17 PogoPlug-E02 ARMV5TE 1.2 GHz Debian Linux TK4- arm_softfloat
hercules port 6 watts"

----------------------------------------------------------------------------------------------

Gee, I can hardly remember how to do this stuff much less what I did several months ago.

1st off it will change the NAND memory in the PogoPlug so that it can boot off of a USB stick.
2nd you have to install Debian Squeeze due to old PogoPlug kernel not supporting something which would allow going straight to current Debian distro.
3rd instead of updating past Squeeze I pulled in some libs (libc was one IIRC)and the new tz (?) package to enable install of new style compressed packages.
4th change TK4- start script as not all programs used in them were installed (system IIRC).

Used putty to get access to Debian on the PogoPlug, mounted USB IDE drive containing TK4-, switched to directory, executed 'mvs' startup script. Shortly logged on to TSO.

Phil
Loading...