Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
» Home
» Features
» News
» Forums
» Classifieds
» Links
» Downloads
Extras
» OS4 Zone
» IRC Network
» AmigaWorld Radio
» Newsfeed
» Top Members
» Amiga Dealers
Information
» About Us
» FAQs
» Advertise
» Polls
» Terms of Service
» Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
18 crawler(s) on-line.
 134 guest(s) on-line.
 1 member(s) on-line.


 A1200

You are an anonymous user.
Register Now!
 A1200:  26 secs ago
 pixie:  9 mins ago
 DiscreetFX:  11 mins ago
 retrofaza:  26 mins ago
 OlafS25:  46 mins ago
 BigD:  47 mins ago
 zErec:  53 mins ago
 amigakit:  56 mins ago
 kolla:  1 hr 40 mins ago
 edwardsjethro:  2 hrs 32 mins ago

/  Forum Index
   /  Amiga OS4 Hardware
      /  Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 Next Page )
PosterThread
wegster 
Summary of Results so far
Posted on 5-Mar-2005 9:16:10
#21 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@thread:
Summary of 'all of the above mess' so far:

uA1-C, G3-750GX 800MHz/1MB cache CPU:
INT READ: ~188 MB/sec>
INT WRITE: ~121 MB/sec
FP READ: ~193 MB/sec
FP WRITE: ~122 MB/sec

A1XE-G3 750FX/512K L2 cache@ 800MHZ, AOS4
INT READ: 199MB/sec
INT WRITE: 124MB/sec
INTMEM: 134MB/sec
FP READ: 199MB/sec
FP WRITE: 125MB/sec
FPMEM: 143MB/sec

XE-G4 7557/256K cache CPU: (AOS4 pre-2, HW fix):
INT READ: ~205MB/second
INT WRITE: ~190MB/second
FP READ: ~209MB/second
FP WRITE: ~415MB/second

XE-G4 7557/256K cache CPU: (Debian 2.4.26 Linux, HW fix):
INT READ: ~206 MB/sec
INT WRITE: ~200 MB/sec
FP READ: ~205 MB/sec
FP WRITE: ~378 MB/sec

A1XE-G4-800Mhz/7451/256K L2 cache, 'unfixed' XE board, AOS4 pre-2:
INT READ: 202MB/sec
INT WRiTE: 210MB/sec
FP READ: 202MB/sec
FP WRITE: 366MB/sec

Apple PowerMac Quicksilver, Dual G4-1GHz/7557 CPU/256L2/2MB L3, PC133 RAM, but n
on SMP benchmark:
INT READ: 394 MB/sec
INT WRITE: 352 MB/sec
FP READ: 397 MB/sec
FP WRITE: 460 MB/sec

FreeBSD 6, 2x667Mhz PIII CPU/256K L2 cache, PC133 Registered ECC RAM, xSeries 45
00R:
INT READ: 465 MB/sec
INT WRITE: 236 MB/sec
FP READ: 465 MB/sec
FP WRITE: 236 MB/sec

3.2GHZ CPU, 1MB L2 cache, DDR2-400 RAM, Linux 2.6.10/gentoo:

INT READ: 3340 MB/sec
INT WRITE: 1610 MB/sec
FP READ: 3700 MB/sec
FP WRITE: 1612 MB/sec

Missing that I'd like for comparisons:

A1SE-G3-600:

A1XE-G4-933/7557

Pegasos 1 w/out April fix:

Peg 1 w/April fix:

Peg 2:

PowerBook, G3 and G4 (compile from source, need to chage the LDFLAGS=-Wl, -O2, -
Wl, -s to LDFLAGS=-Wl, -s in Makefile)

A few other G3/G4 macs.

And ideally but unlikely, A1XE with 750GX CPU module.

Possible conclusions at this point:
1. WTF is up with the INT write speeds on the uA1-C? 121MB/sec versus 190MB/sec?
?! That's more than different grade of RAM can produce.

2. FP writes- I may need to look at the source or do some homework on this one,
that's obviously a huge difference, but may be an optimization on the G4 CPUs.

3. PowerMac with similar hardware, same CPU (but L3 cache enabled)...seems to _s
moke_ the A1s by almost double throughput, with the exception of the XE G4 FP wr
ites. This one strongly would seem to suggest either the A1 memory controller is
n't the greatest, or Apple has done a seriously _amazing_ job with theirs.
Considering the difference here even throwing out the 'L3 boost' on the G4 mac,
I'm thinking it's the former.

4. Is there any reason to think parts of these benchmarks are invalid? if so, w
hy?

DISCLAIMER- These systems are basically hobby and dev platforms at this point, and not in my own opinion meant to be anywhere near 'top end,' basically just 'fast enough,' which I generally think (so far) they are for 'what is needed, now.' None of this should deter anyone in my opinion from buying a uA1 or used XE (I just bought mine myself), but I'd like to have a point of reference when, for example, the XC is brought out.


PS - If no one cares about this type of info, let me know...(and thanks to those who posted their results!)

EDIT- updated w/XE-800 run

Last edited by wegster on 05-Mar-2005 at 11:37 AM.
Last edited by wegster on 05-Mar-2005 at 09:54 AM.
Last edited by wegster on 05-Mar-2005 at 09:43 AM.
Last edited by wegster on 05-Mar-2005 at 09:16 AM.

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
Valiant 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 9:42:54
#22 ]
Super Member
Joined: 21-Oct-2003
Posts: 1109
From: West of Eden, VT USA

@wegster

Quote:
by wegster on 5-Mar-2005 0:44:40

If anyone has an SE, XE, or Peg, please post, even if it's just the average and system specs/cpu version.


For what it's worth here are the numbers from my XE-G4 7441/7451:

ramspeed -b 1
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & WRITING 1 Kb block: 2528.40 Mb/s
INTEGER & WRITING 2 Kb block: 2528.40 Mb/s
INTEGER & WRITING 4 Kb block: 2528.40 Mb/s
INTEGER & WRITING 8 Kb block: 2528.40 Mb/s
INTEGER & WRITING 16 Kb block: 2528.40 Mb/s
INTEGER & WRITING 32 Kb block: 2497.56 Mb/s
INTEGER & WRITING 64 Kb block: 1264.20 Mb/s
INTEGER & WRITING 128 Kb block: 1248.78 Mb/s
INTEGER & WRITING 256 Kb block: 1125.27 Mb/s
INTEGER & WRITING 512 Kb block: 257.29 Mb/s
INTEGER & WRITING 1024 Kb block: 228.06 Mb/s
INTEGER & WRITING 2048 Kb block: 215.35 Mb/s
INTEGER & WRITING 4096 Kb block: 208.55 Mb/s
INTEGER & WRITING 8192 Kb block: 200.39 Mb/s
INTEGER & WRITING 16384 Kb block: 198.64 Mb/s

ramspeed -b 2
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & READING 1 Kb block: 2925.71 Mb/s
INTEGER & READING 2 Kb block: 3011.76 Mb/s
INTEGER & READING 4 Kb block: 2968.12 Mb/s
INTEGER & READING 8 Kb block: 3011.76 Mb/s
INTEGER & READING 16 Kb block: 2968.12 Mb/s
INTEGER & READING 32 Kb block: 2925.71 Mb/s
INTEGER & READING 64 Kb block: 1383.78 Mb/s
INTEGER & READING 128 Kb block: 1383.78 Mb/s
INTEGER & READING 256 Kb block: 1264.20 Mb/s
INTEGER & READING 512 Kb block: 242.65 Mb/s
INTEGER & READING 1024 Kb block: 203.78 Mb/s
INTEGER & READING 2048 Kb block: 202.17 Mb/s
INTEGER & READING 4096 Kb block: 202.17 Mb/s
INTEGER & READING 8192 Kb block: 201.77 Mb/s
INTEGER & READING 16384 Kb block: 202.37 Mb/s

ramspeed -b 4
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & WRITING 1 Kb block: 1969.23 Mb/s
FL-POINT & WRITING 2 Kb block: 1969.23 Mb/s
FL-POINT & WRITING 4 Kb block: 1988.35 Mb/s
FL-POINT & WRITING 8 Kb block: 1932.08 Mb/s
FL-POINT & WRITING 16 Kb block: 1988.35 Mb/s
FL-POINT & WRITING 32 Kb block: 1988.35 Mb/s
FL-POINT & WRITING 64 Kb block: 1765.52 Mb/s
FL-POINT & WRITING 128 Kb block: 1735.59 Mb/s
FL-POINT & WRITING 256 Kb block: 1600.00 Mb/s
FL-POINT & WRITING 512 Kb block: 509.45 Mb/s
FL-POINT & WRITING 1024 Kb block: 400.78 Mb/s
FL-POINT & WRITING 2048 Kb block: 369.68 Mb/s
FL-POINT & WRITING 4096 Kb block: 359.93 Mb/s
FL-POINT & WRITING 8192 Kb block: 352.50 Mb/s
FL-POINT & WRITING 16384 Kb block: 348.89 Mb/s

ramspeed -b 5
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & READING 1 Kb block: 5851.43 Mb/s
FL-POINT & READING 2 Kb block: 5851.43 Mb/s
FL-POINT & READING 4 Kb block: 6023.53 Mb/s
FL-POINT & READING 8 Kb block: 6023.53 Mb/s
FL-POINT & READING 16 Kb block: 6023.53 Mb/s
FL-POINT & READING 32 Kb block: 5851.43 Mb/s
FL-POINT & READING 64 Kb block: 1845.05 Mb/s
FL-POINT & READING 128 Kb block: 1796.49 Mb/s
FL-POINT & READING 256 Kb block: 1678.69 Mb/s
FL-POINT & READING 512 Kb block: 242.65 Mb/s
FL-POINT & READING 1024 Kb block: 205.21 Mb/s
FL-POINT & READING 2048 Kb block: 201.18 Mb/s
FL-POINT & READING 4096 Kb block: 201.77 Mb/s
FL-POINT & READING 8192 Kb block: 201.97 Mb/s
FL-POINT & READING 16384 Kb block: 201.57 Mb/s

_________________
--
-=#Val#=-
Valiant@Camelot


Amiga 1000; Amiga 2000; Amiga 3000T; CD-TV; CD32;
AmigaOne-XE 800Mhz G4;Sam400ep 666Mhz;
AmigaOne X-1000 1.8Ghz PA6T-1682M

 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 9:52:51
#23 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@Valiant

Thanks! So thats the G4-800Mhz, on an 'unfixed' XE board then right? Going to add back into the summary...didn't expect to see any difference from the 'hw fix,' but as details are light, it's worth noting.

Comparing the two XEs:
XE-G4-933Mhz 7557/256K cache CPU: (AOS4 pre-2, 'fixed' XE board):
INT READ: ~205MB/second
INT WRITE: ~190MB/second
FP READ: ~209MB/second
FP WRITE: ~415MB/second

A1XE-G4-800Mhz/7451/256K L2 cache, 'unfixed' XE board:
INT READ: 202MB/sec
INT WRiTE: 210MB/sec
FP READ: 202MB/sec
FP WRITE: 366MB/sec

Pretty close, and within 10%, although _exactly_ for FP writes. Any idea if your RAM is CL2 or CL3?

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 10:05:35
#24 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@thread-
Ok, so tests 3 are NOT just averages, my bad. (Should have read more).

The INT test 1,2 and FP tests 4,5 are basically max possible transfer rate on the hardware, as already stated.

The INT 3 and FP 6 tests:
Quote:

INTmem and FLOATmem, though they are synthetic simulations, but tied closely
with the real world of computing. Each consists of four subtests (Copy, Scale,
Add, Triad), and they measure certain aspects of memory performance. It must be
obvious that any CPU isn't bottomless, so if some data was loaded into its
registers for processing, most likely it will be written back after that, and
probably to another memory location than it was read from. Performance of such
simultaneous read/write operations may tell a lot about a particular chipset
(where controllers driving system and memory buses are located usually), though
a CPU still matters. And it happens often that machines, while showing quite
good linear memory performance, degrade tangibly this way.


So if you guys wouldn't mind, please run the -b 3 and -b 6 tests on your systems and edit the above posts to reflect it, or just post it and I'll add it into my long summary.


_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
Samwel 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 10:53:07
#25 ]
Elite Member
Joined: 7-Apr-2004
Posts: 3404
From: Sweden

@wegster

About the low specs for 750GX.. Can this be because it is not
yet fully supported by OS4?
Maybe the next update of OS4 will have a better result?


/Harry

_________________
/Harry

[SOLD] µA1-C - 750GX 800MHz - 512MB - Antec Aria case

Avatar by HNL_DK!

 Status: Offline
Profile     Report this post  
T_Power 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:04:11
#26 ]
Regular Member
Joined: 8-Sep-2003
Posts: 359
From: Durban, South Africa

@wegster

Here are my results from XE-G3 @ 800MHZ

System: PPC750(G3) -800MHz XE

ramspeed -b1
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & WRITING 1 Kb block: 2925.71 Mb/s
INTEGER & WRITING 2 Kb block: 2925.71 Mb/s
INTEGER & WRITING 4 Kb block: 2925.71 Mb/s
INTEGER & WRITING 8 Kb block: 2968.12 Mb/s
INTEGER & WRITING 16 Kb block: 2925.71 Mb/s
INTEGER & WRITING 32 Kb block: 2968.12 Mb/s
INTEGER & WRITING 64 Kb block: 1484.06 Mb/s
INTEGER & WRITING 128 Kb block: 1494.89 Mb/s
INTEGER & WRITING 256 Kb block: 1484.06 Mb/s
INTEGER & WRITING 512 Kb block: 1432.17 Mb/s
INTEGER & WRITING 1024 Kb block: 123.97 Mb/s
INTEGER & WRITING 2048 Kb block: 123.97 Mb/s
INTEGER & WRITING 4096 Kb block: 123.97 Mb/s
INTEGER & WRITING 8192 Kb block: 123.97 Mb/s
INTEGER & WRITING 16384 Kb block: 123.97 Mb/s

ramspeed -b2
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & READING 1 Kb block: 2968.12 Mb/s
INTEGER & READING 2 Kb block: 2968.12 Mb/s
INTEGER & READING 4 Kb block: 3011.76 Mb/s
INTEGER & READING 8 Kb block: 2968.12 Mb/s
INTEGER & READING 16 Kb block: 3011.76 Mb/s
INTEGER & READING 32 Kb block: 2968.12 Mb/s
INTEGER & READING 64 Kb block: 1575.38 Mb/s
INTEGER & READING 128 Kb block: 1587.60 Mb/s
INTEGER & READING 256 Kb block: 1587.60 Mb/s
INTEGER & READING 512 Kb block: 1539.85 Mb/s
INTEGER & READING 1024 Kb block: 198.64 Mb/s
INTEGER & READING 2048 Kb block: 198.64 Mb/s
INTEGER & READING 4096 Kb block: 198.64 Mb/s
INTEGER & READING 8192 Kb block: 198.64 Mb/s
INTEGER & READING 16384 Kb block: 198.45 Mb/s

ramspeed -b3
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER Copy: 132.39 Mb/s
INTEGER Scale: 131.75 Mb/s
INTEGER Add: 132.36 Mb/s
INTEGER Triad: 138.50 Mb/s
---
INTEGER AVERAGE: 133.75 Mb/s

ramspeed -b4
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & WRITING 1 Kb block: 5688.89 Mb/s
FL-POINT & WRITING 2 Kb block: 5851.43 Mb/s
FL-POINT & WRITING 4 Kb block: 5688.89 Mb/s
FL-POINT & WRITING 8 Kb block: 5851.43 Mb/s
FL-POINT & WRITING 16 Kb block: 5851.43 Mb/s
FL-POINT & WRITING 32 Kb block: 5851.43 Mb/s
FL-POINT & WRITING 64 Kb block: 1969.23 Mb/s
FL-POINT & WRITING 128 Kb block: 1969.23 Mb/s
FL-POINT & WRITING 256 Kb block: 1969.23 Mb/s
FL-POINT & WRITING 512 Kb block: 1878.90 Mb/s
FL-POINT & WRITING 1024 Kb block: 124.73 Mb/s
FL-POINT & WRITING 2048 Kb block: 124.73 Mb/s
FL-POINT & WRITING 4096 Kb block: 124.73 Mb/s
FL-POINT & WRITING 8192 Kb block: 124.80 Mb/s
FL-POINT & WRITING 16384 Kb block: 124.73 Mb/s

ramspeed -b5
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & READING 1 Kb block: 5851.43 Mb/s
FL-POINT & READING 2 Kb block: 6023.53 Mb/s
FL-POINT & READING 4 Kb block: 6023.53 Mb/s
FL-POINT & READING 8 Kb block: 6023.53 Mb/s
FL-POINT & READING 16 Kb block: 6023.53 Mb/s
FL-POINT & READING 32 Kb block: 5851.43 Mb/s
FL-POINT & READING 64 Kb block: 2155.79 Mb/s
FL-POINT & READING 128 Kb block: 2155.79 Mb/s
FL-POINT & READING 256 Kb block: 2133.33 Mb/s
FL-POINT & READING 512 Kb block: 2068.69 Mb/s
FL-POINT & READING 1024 Kb block: 198.64 Mb/s
FL-POINT & READING 2048 Kb block: 198.45 Mb/s
FL-POINT & READING 4096 Kb block: 198.64 Mb/s
FL-POINT & READING 8192 Kb block: 198.64 Mb/s
FL-POINT & READING 16384 Kb block: 198.64 Mb/s

ramspeed -b6
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT Copy: 142.32 Mb/s
FL-POINT Scale: 142.07 Mb/s
FL-POINT Add: 144.23 Mb/s
FL-POINT Triad: 144.29 Mb/s
---
FL-POINT AVERAGE: 143.23 Mb/s


Cheers,
Tim

 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:08:16
#27 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@Samwel
It's possible, but seems odd. I'm sure Hyperion has had uA1s on hand in advance, and I believe the dev pre-release 2 works without 'known issues' on the uA1-C.

It _is_ possible there's something going on here OS related, as I believe a temporary Linux hack was to have the kernel treat the 750GX like one of the known SE/XE G3 CPUs....so I'd love to see the same thing run under Linux, assuming there currently is correct CPU support.

The thing is though, the rest of the numbers look reasonable, at least compared to the XE tests(I'm assuming the FP tests on the uA1-Cs are still doing int tests, so same results). I don't believe there's a whole lot the OS itself does differently regarding memory that's CPU dependent, other than perhaps trying to optimize write block sizes based on the cache size when dealing with large data?

I don't know...I can find out doing a bit of digging in the Linux kernel, or hope someone more into the nuts and bolts of memory access on different hardware can explain a reasonable option here...

FWIW, laptop RAM is generally slower than desktop RAM, but I think this is extreme here with a 63% difference on INT writes...

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:12:27
#28 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@T_Power

EDIT- but left original. It's got to be a 750FX cpu, with 512L2 cache looking at the 'cache hits.' Correct me if I'm wrong anyone...

Is that a 750GX CPU, same as in uA1-C? (if so, how'd you wind up with that combo?)

I'm guessing that's what it is, because it's virtually identical to the uA1-C results. Can you clarify though so there's no question?

If so, then that's interesting...and maybe there is an OS induced limitation at this point with the 750GX CPU.

Any chance you dual boot your system with Linux and can run the same test under Linux? (PM me if you need help getting it to compile...)

Last edited by wegster on 05-Mar-2005 at 11:26 AM.

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
T_Power 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:27:00
#29 ]
Regular Member
Joined: 8-Sep-2003
Posts: 359
From: Durban, South Africa

@wegster

"Is that a 750GX CPU, same as in uA1-C?"

No it's the standard 750FX-G3 that came with the XE.

About Linux on A1, I killed that long time ago... :)

Cheers,
Tim

Last edited by T_Power on 05-Mar-2005 at 11:29 AM.

 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:40:24
#30 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@T_Power
Tim- thanks. Must've been typing at same time, I reazlied it must be a FX after writing that and edited the other post

So...this is strange then. Both the 750FX and 750GX show slower writes, both in the uA1-C AND in the XE board, so it's not a board/chipset difference.

Hmm. Any chance anyone is running a 750FX or GX under Linux and can build and test ramspeed there?

It may be possible it's a compiler optimization bug for that specific processor set. The source is simple, I'll see if I can't get it to build under the slightly newer version (version on os4depot is 2.3.9, current is 2.3.1), but I expect it was probably the same version of gcc that compiled it...it _might_ be worth trying a non-optimized version in case of a gcc issue. Open for other ideas of course..?

Last edited by wegster on 05-Mar-2005 at 11:47 AM.

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
mlehto 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 11:50:16
#31 ]
Super Member
Joined: 4-Dec-2004
Posts: 1006
From: Unknown

@wegster


This is intresting...

Ppl shouls test drbombcrater SetA1 tool (mem tweaker).

You have memory stick from kgrach, officially working as good as possible ??

I got similar results with my SE with ramspeed than ppl here with uA1.



 Status: Offline
Profile     Report this post  
wegster 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 12:01:57
#32 ]
Elite Member
Joined: 29-Nov-2004
Posts: 8554
From: RTP, NC USA

@mlehto
I actually PMed DrBombCrater asking him where to grab that from...is it publicly available?

Which SE do you have? CPU/speed...you're seeing the 'slow writes' then?

The source for ramspeed is clean C, compiles fine, but I need to figure out how to get gcc happy on my system and what's special RE: AOS builds..

Last edited by wegster on 05-Mar-2005 at 12:11 PM.

_________________
Are we not done with the same silly arguments and flames yet??!

 Status: Offline
Profile     Report this post  
Zorro 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 12:23:58
#33 ]
Super Member
Joined: 30-Apr-2003
Posts: 1081
From: Italy

@wegster

Wow ! These are a lot of numbers...


Quote:
It _is_ possible there's something going on here OS related, as I believe a temporary Linux hack was to have the kernel treat the 750GX like one of the known SE/XE G3 CPUs....so I'd love to see the same thing run under Linux, assuming there currently is correct CPU support.


Well, GX support AFAIK has yet some problems on both OS4 and Linux ...

Last edited by Zorro on 05-Mar-2005 at 12:26 PM.

_________________
-------------------------------
AmigaOS, the last hope...

 Status: Offline
Profile     Report this post  
itix 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 12:40:27
#34 ]
Elite Member
Joined: 22-Dec-2004
Posts: 3398
From: Freedom world

@wegster

Pegasos 2

CPU: 7447 1000MHz (G4)
RAM: 768MB

And here are results (I ran OS4 memspeed binary):

Ram Disk:> os4emu ramspeed -b 1
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & WRITING 1 Kb block: 3200.00 Mb/s
INTEGER & WRITING 2 Kb block: 3200.00 Mb/s
INTEGER & WRITING 4 Kb block: 3250.79 Mb/s
INTEGER & WRITING 8 Kb block: 3200.00 Mb/s
INTEGER & WRITING 16 Kb block: 3250.79 Mb/s
INTEGER & WRITING 32 Kb block: 3200.00 Mb/s
INTEGER & WRITING 64 Kb block: 1638.40 Mb/s
INTEGER & WRITING 128 Kb block: 1638.40 Mb/s
INTEGER & WRITING 256 Kb block: 1625.40 Mb/s
INTEGER & WRITING 512 Kb block: 1505.88 Mb/s
INTEGER & WRITING 1024 Kb block: 275.64 Mb/s
INTEGER & WRITING 2048 Kb block: 240.66 Mb/s
INTEGER & WRITING 4096 Kb block: 238.14 Mb/s
INTEGER & WRITING 8192 Kb block: 236.22 Mb/s
INTEGER & WRITING 16384 Kb block: 235.94 Mb/s

Ram Disk:> os4emu ramspeed -b 2
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER & READING 1 Kb block: 3531.03 Mb/s
INTEGER & READING 2 Kb block: 3723.64 Mb/s
INTEGER & READING 4 Kb block: 3792.59 Mb/s
INTEGER & READING 8 Kb block: 3723.64 Mb/s
INTEGER & READING 16 Kb block: 3792.59 Mb/s
INTEGER & READING 32 Kb block: 3792.59 Mb/s
INTEGER & READING 64 Kb block: 1780.87 Mb/s
INTEGER & READING 128 Kb block: 1780.87 Mb/s
INTEGER & READING 256 Kb block: 1765.52 Mb/s
INTEGER & READING 512 Kb block: 1563.36 Mb/s
INTEGER & READING 1024 Kb block: 269.12 Mb/s
INTEGER & READING 2048 Kb block: 229.85 Mb/s
INTEGER & READING 4096 Kb block: 221.89 Mb/s
INTEGER & READING 8192 Kb block: 220.93 Mb/s
INTEGER & READING 16384 Kb block: 226.80 Mb/s

Ram Disk:> os4emu ramspeed -b 3
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

INTEGER Copy: 224.07 Mb/s
INTEGER Scale: 222.85 Mb/s
INTEGER Add: 192.66 Mb/s
INTEGER Triad: 190.10 Mb/s
---
INTEGER AVERAGE: 207.42 Mb/s

Ram Disk:> os4emu ramspeed -b 4
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & WRITING 1 Kb block: 2497.56 Mb/s
FL-POINT & WRITING 2 Kb block: 2528.40 Mb/s
FL-POINT & WRITING 4 Kb block: 948.15 Mb/s
FL-POINT & WRITING 8 Kb block: 948.15 Mb/s
FL-POINT & WRITING 16 Kb block: 948.15 Mb/s
FL-POINT & WRITING 32 Kb block: 948.15 Mb/s
FL-POINT & WRITING 64 Kb block: 948.15 Mb/s
FL-POINT & WRITING 128 Kb block: 943.78 Mb/s
FL-POINT & WRITING 256 Kb block: 948.15 Mb/s
FL-POINT & WRITING 512 Kb block: 890.43 Mb/s
FL-POINT & WRITING 1024 Kb block: 345.95 Mb/s
FL-POINT & WRITING 2048 Kb block: 323.03 Mb/s
FL-POINT & WRITING 4096 Kb block: 321.00 Mb/s
FL-POINT & WRITING 8192 Kb block: 321.51 Mb/s
FL-POINT & WRITING 16384 Kb block: 321.00 Mb/s

Ram Disk:> os4emu ramspeed -b 5
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT & READING 1 Kb block: 7314.29 Mb/s
FL-POINT & READING 2 Kb block: 7585.19 Mb/s
FL-POINT & READING 4 Kb block: 1517.04 Mb/s
FL-POINT & READING 8 Kb block: 1517.04 Mb/s
FL-POINT & READING 16 Kb block: 1505.88 Mb/s
FL-POINT & READING 32 Kb block: 1517.04 Mb/s
FL-POINT & READING 64 Kb block: 1163.64 Mb/s
FL-POINT & READING 128 Kb block: 1163.64 Mb/s
FL-POINT & READING 256 Kb block: 1163.64 Mb/s
FL-POINT & READING 512 Kb block: 1077.89 Mb/s
FL-POINT & READING 1024 Kb block: 263.58 Mb/s
FL-POINT & READING 2048 Kb block: 230.37 Mb/s
FL-POINT & READING 4096 Kb block: 227.05 Mb/s
FL-POINT & READING 8192 Kb block: 226.80 Mb/s
FL-POINT & READING 16384 Kb block: 227.05 Mb/s

Ram Disk:> os4emu ramspeed -b 6
OS4Emu by Ilkka Lehtoranta. All rights reserved.
Release 1.6. Command version 1.6.
RAMspeed (UNIX) v2.3.0 by Rhett M. Hollander (Alasir Enterprises), 2002-04

4Gb per pass mode

FL-POINT Copy: 228.95 Mb/s
FL-POINT Scale: 227.68 Mb/s
FL-POINT Add: 200.00 Mb/s
FL-POINT Triad: 194.18 Mb/s
---
FL-POINT AVERAGE: 212.71 Mb/s

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

 Status: Offline
Profile     Report this post  
DrBombcrater 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 15:09:50
#35 ]
Super Member
Joined: 6-Feb-2004
Posts: 1382
From: UK

@wegster

Quote:
I actually PMed DrBombCrater asking him where to grab that from...is it publicly available?

Sorry, I didn't notice your PM

I pulled the public version of SetA1 because it had a couple of serious bugs. The new version is almost done and will be out next week.

_________________
Who do you serve, and who do you trust? - Galen

 Status: Offline
Profile     Report this post  
JKD 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 15:53:10
#36 ]
Regular Member
Joined: 17-Aug-2003
Posts: 210
From: South of Heaven

@itix

I made a native compile of that utility (well, I downloaded the source and typed make!) and it shows different results on the Peg2. I'll post later. I think it maybe shows there are different optimizatinos between the versions and/or os4emu somehow has an influence (which would be weird)

Steve

 Status: Offline
Profile     Report this post  
minator 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 15:57:13
#37 ]
Cult Member
Joined: 23-Mar-2004
Posts: 989
From: Cambridge

@wegster

Quote:
1. WTF is up with the INT write speeds on the uA1-C? 121MB/sec versus 190MB/sec??! That's more than different grade of RAM can produce.


I did some memory tests a couple of years back on the Peg 1 and a G3 iBook and these are pretty similar.

It appears the G3 memory write speed is not that great.
G4 improves in both areas, especially writes.

There appear to be 2 Apple memory controllers, one sucks badly (was in the G3 iBook) and a second much better vesion in the G4 tower and my PowerBook, ArticiaS reads slower than the sucky Apple one but writes faster.


I'll attempt to interpret these to explain what's going on:

As with memny memory test it's not actually testing the memory directly so these values are impacted by the cache speed. If you want the memory speed only you need to write a test which only uses the internal registers (preferably FP or Altivec).

That said these figures are quite useful as I'm writing a modular synth and these tell me the working set I can use at any one time.

From the "INTEGER & WRITING" figures below:

1kb - 32Kb blocks: (3.6-3.9 GB/second) - looks like this is hitting the L1 cache
64kb - 256kb (2 GB / second) - looks like this is hitting the L2 cache
512kb - 1024kb (420 MB - 1Gb / second) - starting to run out of L2.
4096kb upwards (374 MB / second) - cache is having a small effect - this is the RAM speed.

My results PowerBook G4 (7447A 1.33GHz):

./ramspeed -b 1

INTEGER & WRITING 1 Kb block: 3684.47 Mb/s
INTEGER & WRITING 2 Kb block: 3941.52 Mb/s
INTEGER & WRITING 4 Kb block: 3830.35 Mb/s
INTEGER & WRITING 8 Kb block: 3975.71 Mb/s
INTEGER & WRITING 16 Kb block: 3830.78 Mb/s
INTEGER & WRITING 32 Kb block: 3739.87 Mb/s
INTEGER & WRITING 64 Kb block: 2004.19 Mb/s
INTEGER & WRITING 128 Kb block: 2007.32 Mb/s
INTEGER & WRITING 256 Kb block: 1996.97 Mb/s
INTEGER & WRITING 512 Kb block: 1045.27 Mb/s
INTEGER & WRITING 1024 Kb block: 420.61 Mb/s
INTEGER & WRITING 2048 Kb block: 378.21 Mb/s
INTEGER & WRITING 4096 Kb block: 374.96 Mb/s
INTEGER & WRITING 8192 Kb block: 374.20 Mb/s
INTEGER & WRITING 16384 Kb block: 374.39 Mb/s



./ramspeed -b 2

INTEGER & READING 1 Kb block: 4292.33 Mb/s
INTEGER & READING 2 Kb block: 4674.13 Mb/s
INTEGER & READING 4 Kb block: 4683.27 Mb/s
INTEGER & READING 8 Kb block: 4365.84 Mb/s
INTEGER & READING 16 Kb block: 4597.62 Mb/s
INTEGER & READING 32 Kb block: 4248.39 Mb/s
INTEGER & READING 64 Kb block: 2172.73 Mb/s
INTEGER & READING 128 Kb block: 2159.78 Mb/s
INTEGER & READING 256 Kb block: 1627.83 Mb/s
INTEGER & READING 512 Kb block: 847.00 Mb/s
INTEGER & READING 1024 Kb block: 398.08 Mb/s
INTEGER & READING 2048 Kb block: 348.73 Mb/s
INTEGER & READING 4096 Kb block: 344.38 Mb/s
INTEGER & READING 8192 Kb block: 343.94 Mb/s
INTEGER & READING 16384 Kb block: 340.01 Mb/s



./ramspeed -b 3

INTEGER Copy: 385.95 Mb/s
INTEGER Scale: 384.92 Mb/s
INTEGER Add: 392.12 Mb/s
INTEGER Triad: 370.35 Mb/s
---
INTEGER AVERAGE: 383.34 Mb/s



./ramspeed -b 4

FL-POINT & WRITING 1 Kb block: 3086.20 Mb/s
FL-POINT & WRITING 2 Kb block: 2966.10 Mb/s
FL-POINT & WRITING 4 Kb block: 3019.00 Mb/s
FL-POINT & WRITING 8 Kb block: 3147.40 Mb/s
FL-POINT & WRITING 16 Kb block: 3059.98 Mb/s
FL-POINT & WRITING 32 Kb block: 3040.37 Mb/s
FL-POINT & WRITING 64 Kb block: 2543.82 Mb/s
FL-POINT & WRITING 128 Kb block: 2557.73 Mb/s
FL-POINT & WRITING 256 Kb block: 1850.67 Mb/s
FL-POINT & WRITING 512 Kb block: 1023.73 Mb/s
FL-POINT & WRITING 1024 Kb block: 546.63 Mb/s
FL-POINT & WRITING 2048 Kb block: 474.73 Mb/s
FL-POINT & WRITING 4096 Kb block: 468.30 Mb/s
FL-POINT & WRITING 8192 Kb block: 469.32 Mb/s
FL-POINT & WRITING 16384 Kb block: 464.35 Mb/s



./ramspeed -b 5

FL-POINT & READING 1 Kb block: 8584.88 Mb/s
FL-POINT & READING 2 Kb block: 8124.13 Mb/s
FL-POINT & READING 4 Kb block: 8905.01 Mb/s
FL-POINT & READING 8 Kb block: 8915.55 Mb/s
FL-POINT & READING 16 Kb block: 9082.04 Mb/s
FL-POINT & READING 32 Kb block: 9035.10 Mb/s
FL-POINT & READING 64 Kb block: 2825.28 Mb/s
FL-POINT & READING 128 Kb block: 2817.86 Mb/s
FL-POINT & READING 256 Kb block: 2167.45 Mb/s
FL-POINT & READING 512 Kb block: 895.08 Mb/s
FL-POINT & READING 1024 Kb block: 407.37 Mb/s
FL-POINT & READING 2048 Kb block: 346.92 Mb/s
FL-POINT & READING 4096 Kb block: 344.70 Mb/s
FL-POINT & READING 8192 Kb block: 344.12 Mb/s
FL-POINT & READING 16384 Kb block: 344.44 Mb/s



./ramspeed -b 6

FL-POINT Copy: 428.99 Mb/s
FL-POINT Scale: 430.94 Mb/s
FL-POINT Add: 449.87 Mb/s
FL-POINT Triad: 450.75 Mb/s
---
FL-POINT AVERAGE: 440.13 Mb/s


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

Cache (max)
Reading 9 GB / second
Writing 3.975 GB / Second

Memory
Reading 344 MB / second
Writing 467 MB / second



_________________
Whyzzat?

 Status: Offline
Profile     Report this post  
opi 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 15:58:39
#38 ]
Team Member
Joined: 2-Mar-2005
Posts: 2752
From: Poland

@wegster

AMD Duron with 266 DDR running GNU/Debian:

fretka:~/ramspeed-2.3.1# ./ramspeed -b 1
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

INTEGER & WRITING 1 Kb block: 7429.98 Mb/s
INTEGER & WRITING 2 Kb block: 7459.04 Mb/s
INTEGER & WRITING 4 Kb block: 7303.42 Mb/s
INTEGER & WRITING 8 Kb block: 7385.49 Mb/s
INTEGER & WRITING 16 Kb block: 7399.19 Mb/s
INTEGER & WRITING 32 Kb block: 7438.37 Mb/s
INTEGER & WRITING 64 Kb block: 7364.59 Mb/s
INTEGER & WRITING 128 Kb block: 2076.84 Mb/s
INTEGER & WRITING 256 Kb block: 352.61 Mb/s
INTEGER & WRITING 512 Kb block: 357.71 Mb/s
INTEGER & WRITING 1024 Kb block: 363.82 Mb/s
INTEGER & WRITING 2048 Kb block: 361.20 Mb/s
INTEGER & WRITING 4096 Kb block: 362.50 Mb/s
INTEGER & WRITING 8192 Kb block: 362.22 Mb/s
INTEGER & WRITING 16384 Kb block: 359.58 Mb/s

fretka:~/ramspeed-2.3.1# ./ramspeed -b 4
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

FL-POINT & WRITING 1 Kb block: 7907.92 Mb/s
FL-POINT & WRITING 2 Kb block: 7925.33 Mb/s
FL-POINT & WRITING 4 Kb block: 7916.37 Mb/s
FL-POINT & WRITING 8 Kb block: 7795.46 Mb/s
FL-POINT & WRITING 16 Kb block: 7856.95 Mb/s
FL-POINT & WRITING 32 Kb block: 7886.72 Mb/s
FL-POINT & WRITING 64 Kb block: 7868.46 Mb/s
FL-POINT & WRITING 128 Kb block: 2591.17 Mb/s
FL-POINT & WRITING 256 Kb block: 442.08 Mb/s
FL-POINT & WRITING 512 Kb block: 449.97 Mb/s
FL-POINT & WRITING 1024 Kb block: 446.68 Mb/s
FL-POINT & WRITING 2048 Kb block: 441.77 Mb/s
FL-POINT & WRITING 4096 Kb block: 442.78 Mb/s
FL-POINT & WRITING 8192 Kb block: 442.37 Mb/s
FL-POINT & WRITING 16384 Kb block: 442.77 Mb/s

fretka:~/ramspeed-2.3.1# ./ramspeed -b 6
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

FL-POINT Copy: 452.23 Mb/s
FL-POINT Scale: 472.91 Mb/s
FL-POINT Add: 512.01 Mb/s
FL-POINT Triad: 508.00 Mb/s
---
FL-POINT AVERAGE: 486.29 Mb/s

fretka:~/ramspeed-2.3.1# uname -a
Linux fretka 2.6.8-1-386 #1 Thu Nov 11 12:18:43 EST 2004 i686 GNU/Linux

fretka:~/ramspeed-2.3.1# cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 8
model name : AMD Duron(tm) Processor
stepping : 1
cpu MHz : 1045.758
cache size : 64 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow
bogomips : 2076.67

Last edited by opi on 05-Mar-2005 at 03:58 PM.

_________________
OpenWindows Initiative. Port PS3 hardware to bananas. For free. Join today and receive expired $50 cupon from AI!

 Status: Offline
Profile     Report this post  
itix 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 16:12:36
#39 ]
Elite Member
Joined: 22-Dec-2004
Posts: 3398
From: Freedom world

@JKD

Interesting!

Ok, here are my results using native compile:

(Peg2, 7447 1GHz)

Ram Disk:ramspeed-2.3.1> make
gcc -Wall -O2 -c -o ramspeed.o ramspeed.c
gcc -Wall -O2 -c -o intmark.o intmark.c
gcc -Wall -O2 -c -o intmem.o intmem.c
gcc -Wall -O2 -c -o fltmark.o fltmark.c
gcc -Wall -O2 -c -o fltmem.o fltmem.c
gcc -Wl,-O2 -Wl,-s -o ramspeed ramspeed.o\
intmark.o intmem.o fltmark.o fltmem.o
Ram Disk:ramspeed-2.3.1> ramspeed -b 1
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

INTEGER & WRITING 1 Kb block: 3215.20 Mb/s
INTEGER & WRITING 2 Kb block: 3233.58 Mb/s
INTEGER & WRITING 4 Kb block: 3237.82 Mb/s
INTEGER & WRITING 8 Kb block: 3229.93 Mb/s
INTEGER & WRITING 16 Kb block: 3240.14 Mb/s
INTEGER & WRITING 32 Kb block: 3209.32 Mb/s
INTEGER & WRITING 64 Kb block: 1589.73 Mb/s
INTEGER & WRITING 128 Kb block: 1588.98 Mb/s
INTEGER & WRITING 256 Kb block: 1580.93 Mb/s
INTEGER & WRITING 512 Kb block: 1454.09 Mb/s
INTEGER & WRITING 1024 Kb block: 505.33 Mb/s
INTEGER & WRITING 2048 Kb block: 470.28 Mb/s
INTEGER & WRITING 4096 Kb block: 464.35 Mb/s
INTEGER & WRITING 8192 Kb block: 464.38 Mb/s
INTEGER & WRITING 16384 Kb block: 464.80 Mb/s
Ram Disk:ramspeed-2.3.1> ramspeed -b 2
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

INTEGER & READING 1 Kb block: 3704.41 Mb/s
INTEGER & READING 2 Kb block: 3741.35 Mb/s
INTEGER & READING 4 Kb block: 3764.05 Mb/s
INTEGER & READING 8 Kb block: 3772.80 Mb/s
INTEGER & READING 16 Kb block: 3761.38 Mb/s
INTEGER & READING 32 Kb block: 3739.05 Mb/s
INTEGER & READING 64 Kb block: 1777.16 Mb/s
INTEGER & READING 128 Kb block: 1774.09 Mb/s
INTEGER & READING 256 Kb block: 1767.18 Mb/s
INTEGER & READING 512 Kb block: 1592.13 Mb/s
INTEGER & READING 1024 Kb block: 268.62 Mb/s
INTEGER & READING 2048 Kb block: 228.87 Mb/s
INTEGER & READING 4096 Kb block: 225.56 Mb/s
INTEGER & READING 8192 Kb block: 225.63 Mb/s
INTEGER & READING 16384 Kb block: 225.61 Mb/s
Ram Disk:ramspeed-2.3.1> ramspeed -b 3
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

INTEGER Copy: 237.46 Mb/s
INTEGER Scale: 227.35 Mb/s
INTEGER Add: 197.28 Mb/s
INTEGER Triad: 200.65 Mb/s
---
INTEGER AVERAGE: 215.68 Mb/s
Ram Disk:ramspeed-2.3.1> ramspeed -b 4
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

FL-POINT & WRITING 1 Kb block: 947.12 Mb/s
FL-POINT & WRITING 2 Kb block: 946.31 Mb/s
FL-POINT & WRITING 4 Kb block: 947.35 Mb/s
FL-POINT & WRITING 8 Kb block: 946.03 Mb/s
FL-POINT & WRITING 16 Kb block: 945.39 Mb/s
FL-POINT & WRITING 32 Kb block: 946.74 Mb/s
FL-POINT & WRITING 64 Kb block: 946.75 Mb/s
FL-POINT & WRITING 128 Kb block: 945.60 Mb/s
FL-POINT & WRITING 256 Kb block: 943.49 Mb/s
FL-POINT & WRITING 512 Kb block: 888.79 Mb/s
FL-POINT & WRITING 1024 Kb block: 500.46 Mb/s
FL-POINT & WRITING 2048 Kb block: 471.35 Mb/s
FL-POINT & WRITING 4096 Kb block: 464.72 Mb/s
FL-POINT & WRITING 8192 Kb block: 465.33 Mb/s
FL-POINT & WRITING 16384 Kb block: 464.84 Mb/s
Ram Disk:ramspeed-2.3.1> ramspeed -b 5
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

FL-POINT & READING 1 Kb block: 1511.37 Mb/s
FL-POINT & READING 2 Kb block: 1515.32 Mb/s
FL-POINT & READING 4 Kb block: 1512.77 Mb/s
FL-POINT & READING 8 Kb block: 1514.41 Mb/s
FL-POINT & READING 16 Kb block: 1514.56 Mb/s
FL-POINT & READING 32 Kb block: 1508.46 Mb/s
FL-POINT & READING 64 Kb block: 1164.39 Mb/s
FL-POINT & READING 128 Kb block: 1162.77 Mb/s
FL-POINT & READING 256 Kb block: 1157.30 Mb/s
FL-POINT & READING 512 Kb block: 1065.39 Mb/s
FL-POINT & READING 1024 Kb block: 262.41 Mb/s
FL-POINT & READING 2048 Kb block: 229.30 Mb/s
FL-POINT & READING 4096 Kb block: 226.23 Mb/s
FL-POINT & READING 8192 Kb block: 226.37 Mb/s
FL-POINT & READING 16384 Kb block: 225.99 Mb/s
Ram Disk:ramspeed-2.3.1> ramspeed -b 6
RAMspeed (UNIX) v2.3.1 by Rhett M. Hollander (Alasir Enterprises), 2002-05

4Gb per pass mode

FL-POINT Copy: 223.14 Mb/s
FL-POINT Scale: 223.01 Mb/s
FL-POINT Add: 201.62 Mb/s
FL-POINT Triad: 202.89 Mb/s
---
FL-POINT AVERAGE: 212.66 Mb/s


I dont think it is OS4EMU issue... hmm...

(edit: removed empty lines below)

Last edited by itix on 05-Mar-2005 at 04:16 PM.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

 Status: Offline
Profile     Report this post  
mlehto 
Re: Can someone with a uA1-C (and/or Peg) run 'ramspeed' for me?
Posted on 5-Mar-2005 17:17:16
#40 ]
Super Member
Joined: 4-Dec-2004
Posts: 1006
From: Unknown

@wegster

No, I got it from DWolfman ...

drbombcrater SetA1 make things different ... well done !-) (more tools like this)



I have 666/133 clocked SE. It is same "incarnation" as drbombcrater has, first public release. Memory stick is Kingston 512 MB reg ECC CL3.

Some ppl drive their SE's like 733/133. Altoughht I have Zallmann NB cooler, I like too much my SE

I definetly see slow writes and reads



I use memory speed tool made by Harry Sintonen (known better as Piru) and it gives "better" results. Ie results wary more. But it is not publically avail yet.

With this very tool I got 120MB/s and 210 MB/s results, They got better with SetA1 tool, 140 MB/s and 260 MB/s.

My friend got 356 MB/s 32/64 bit read and 374 MB/s 32bit write and 625 MB/s 64 bit write with XE (1066/133 G4), with exactly same type of memory stick. So there should be something different between SE and XE.

XE didn't responce to SetA1 much. Why, don't know.

If I remember correctly, I got ramspeed results 120 MB/s and 160 MB/s.

You can see, that at least with this tool memoryspeeds get up. It should be insane, if nothing don't make any differ. Ppl in here have at least 666 G3 CXe, 800 FX/GX processors, varioius G4:s 800/933/1066, all popultaed with different type of memory sticks without any big differencies ?? I didn't want to read all results thought too carefully...

These are after all "only" benchmarks. But it should be nice to know, where is bottleneck. OS itself, ArticiaS or how things are "prepared" by uboot.

Anyway I feel that Piru's tool works better. Not because better results, but because it give results in larger scale.



Doh ... Have anyone (!!!!) CL2 unreg non ECC memory stick. It could be intresting to know results. Anyway it have to be tweaked with SetA1, without results may be just same, I presume. In XE machines and up this kind of stick may (or may not...) work. At least as single stick.



This test made by itix was somehow intresting, at least you cannot say anything sure about emulated tests.

And this test made with apple was cood also.

Both with PPC at least.



I didn't test this, but if you compile ramspeed yourself, compile it with -O2 switch and proper prosessor switch enabled.


This was long one. Hope that I said something reasonable ...

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 Next Page )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle