Test Your Setup Performance...

General Discussion
21

i don’t know what you’re measuring there.
but there’s no way you get this performance with that system in this test.

there is also no relation between cpuloadbench.rns result and the test.rns ones.
the cpuloadbench.rns is WAY more heavy on the CPU and you only get a 2.5% difference?
there went something pretty wrong IMHO.

@timo
very nice results, given that your 2ghz dothan isn’t overclocked even!
just imagine what this little puppy would do if you’d take him up to ~2,5-2,7Ghz like most other ppl do that are serious about overclocking.

22

yes… its related to pci and pci express bus

23

I suspect the performance would be even greater with a PCI based soundcard (such as an Emu 1616m) instead of a Firewire one. My laptop has a PCI-expresscard instead of PCMCIA so that wasn’t an option.

Trying to find a way around the bios so I can overclock it. Fujitsu have locked/hidden absolutely everything in it other than the startup boot device config. <_<

24

i’m not too familiar with notebook overclocking, but i suppose even if the BIOS would allow to adjust clocks and voltages, the results would be pretty lousy because of the limited cooling abilities notebooks offer.

this adapter enables you to run dothans on Asus S478 boards, just like my P4C800E-Dlx, but the problem is that my watercooling is not compatible with the socket479 and i’d have to modify it myself to make it fit. so most probably i’d end up with a dead motherboard and/or CPU, so i skipped that idea. ;)
but with that adapter other people tend to get 2,6ghz on air out of a dothan 2.0 or 2,13.
on water even 3ghz+ which would be really awesome, performance wise.
these CPU are true FPU monsters, in contrast to the desktop P4s.
but then again, they lack hyperthreading… which doesn’t help with renoise, but is still something i’d personally not want to miss anymore when using several ordinary applications at once.

25

I’ve seen evidence of people oc’ing venice core A64 3000+ +50% on air (HardOCP).

Wish I could run the test, but I don’t have the full version of renoise (asio)… YET.

I’ve also come across some data that suggests that with a64’s, tight memory timings and speed are not as important for a good oc as with other cpus (some DFI forum).

26

A rock stable 2.4GHz o/clock has been got from 2.1GHz PM @ 533MHz laptops, without much overhead (heat/noise). Am kicking myself I can’t access the hidden BIOS settings on my own laptop.

…As for PentiumMs for the future, just imagine when dual-core, 4mb-L2 64-bit pentium Ms get here… :w00t: :yeah:

That’s two generations away, though (“Merom”). ;)

Hope they’re as quiet/cool as the current PMs, but that’ll be a task and a half (ie. I doubt it).

27

well, 50% overclock on A64 3000+ are “just” 2,7Ghz.
it sure is a nice CPU then and the 3000+ venice is known to clock very good, but still i’d prefer a dothan running at 2,7Ghz (or even higher like this or even this, which is easily possible on water) as it performs better clock-by-clock compared to the A64. (see superpi results on various forums → FPU heavy, same goes for renoise as it seems).

hehe, yeah i personally can’t wait, too ! :)
and it sure won’t be expensive at all when it’s out ;)

28

first of all, i’d like to say that i’m not keen on an INTEL vs AMD or whatever kind of “fanboy flamewar”.
i’m not biased towards this and price/performance-wise i always recommend AMD systems, which has also happened in this forum several times before.
however, what you say is simply not true, because the PentiumM’s FPU, respectively the one of the Dothan PM, works factually more efficient than those of the San Diego, Venice or virtually any other K8-A64 core known to the enduser market.
please take a look at the chart linked below, which represents the platform-spanning, gathered results of a program called “SuperPI” which calculates the number PI with a given amount of decimal places in a system performance depending amount of time.
the lower the time for the (1M) test, the faster your system’s ability to calculate floating point operations.
please compare clockspeeds of dothans & K8 CPUs and their time needed for the same calculation.

TOP 50 SuperPI results (gathered by the users of one certain forum)

29

that merely proves that the pro53 synth is less dependant on RAM speed than superpi, but is no indication for your initial statement, which i do not agree with, to be true.
clock-by-clock the PMs performs better in ~90% of all applications available for x86 machines. be it games, content creation software, video rendering, audio or office related applications. there has also been a similar benchmarking thread not long ago on the ableton live forums, with a lot more participation than here. dothans yield the best results there, too.
so please read some reviews of various magazines/hardwareenthusiast/whateversites and you will figure.
and btw, you should’ve noticed that in the list linked in my previous post, the 2nd ranked entry, a dothan, has a way lower CPU and FSB/RAM clock than the 3rd placed FX-57 1000$ CPU, based on the San Diego core.

whatever… i fear this discussion is going nowhere and should therefore halt here, before this thread loses its original intention to collect renoise benchmark results.
so keep’em coming!

30

A64 3000+ with slight oc.
Epox 9npaj
1 gig pc3200 ram.
200 gig hd.
gforce 7800 gt.
creative xfi. (so sue me I play a lot of games. I’ll get a different one if I need it).

@44.1

[File]
CPU_LOAD_BENCH.RNS

[20 ms]
CPU peaks @ 67%

[12 ms]
CPU peaks @ 72.4%

[06 ms]
CPU peaks @ 75.5%

[03 ms]
CPU peaks @ 84.5% (/w crackles)

31

regarding the above review, which takes all performance related factors, be it soft- or hardware-wise, of both systems into detailed account, i presume we finally have the empirical, highly qualified and scientifically correct proof concerning this matter.
respect, sherlock.

  • AMD fanboy detected.
  • conversation halted.
  • re-insert coin.
32

A64 4000+ (2400MHz)
Asus A8N-E
2 gig ram.
Delta 1010

@44.1

[File]
CPU_LOAD_BENCH.RNS

[17 ms]
CPU peaks @ 59%

[11ms]
CPU peaks @ 62%

[08 ms]
CPU peaks @ 64%

[05 ms]
CPU peaks @ 69%

I will give o/c this machine today, and see hows it affect performance.

33

A64 4000+ (oc’d to 2736MHz)
Asus A8N-E
2 gig ram.
Delta 1010

@44.1

[File]
CPU_LOAD_BENCH.RNS

[17 ms]
CPU peaks @ 51%%

[11ms]
CPU peaks @ 54%

[08 ms]
CPU peaks @ 56%

[05 ms]
CPU peaks @ 60%

In general, Im very happy to my new system. Fast and reliably (so far).

34

Thought I should test these again after I changed machine.
Only tried my current settings though.

Intel Core 2 Duo 6700 - OC’ed to 3GHz
1066 FSB
1GB DDR2-8500
M-audio Audiophile 2496
Asus 7900GT @ 1920x1200

[CPU Load]

35%, 128 Samples/3ms latency @ 44.1kHz

[test.rns]

26%, 128 Samples/3ms latency @ 44.1kHz

35

I haven’t ran the CPU_BENCH file since I had my socket A 2500+ which was barely able to play it at most
rates & buffersizes, but I’m pleased to see how much things have improved :)

System spec:

Opteron 170 @ 2.75GHz on 1.45v
DFI Lanparty NF4 SLI-DR Expert
2x512MB Corsair XMS3200XLPT @ 275MHz (2.5-3-3-6-1T)
M-Audio Delta 1010

No Dual core optimiser installed

test.rns:

Frequency: 96KHz

Buffersize: 64-128
Latency: ‘0’-1ms
CPU Load Very jumpy from 60-99%
Very weird ‘lofimat’ sound, less apparent with 128 buffer

Buffersize: 256,384,512,768
Latency: 2-8ms
CPU Load: 92%, less jumpy
Still pops, 768 buffer is more listenable

896 buffersize hangs on playback.

Buffersize: 1024
Latency: 10ms
CPU Load: 76%
Dropouts stop after 30 seconds or so.

Buffersize: 1280
Latency: 13ms
CPU Load: 76%
No dropouts.

Buffersize: 2048
Latency: 21ms
CPU Load: 75%

So, no major improvement on CPU load going beyond 1280 buffersize at 96KHz.

Frequency: 44.1KHz

Buffersize: 64
Latency: 2ms
CPU Load: 43%

Buffersize: 256
Latency: 6ms
CPU Load: 41%

Buffersize: 2048
Latency: 47ms
CPU Load: 37%

So, as expected, not much improvement between even 64-2048 buffersizes at 44.1KHz. Very glad to not notice any dropouts.

CPU_LOAD_BENCH.rns

Frequency: 96KHz

with buffers from 64-768

Latency: '0-8’ms
CPU Load 99.9%
It plays without stopping or grinding to a halt (probably due to core management), but a ‘timestretched’ effect occurs due to dropouts. This improves as the buffer size is increased up to 768, where the majority of it plays smoothly.

Buffersize: 896
Hangs on playback once again!

Buffersize: 1024
Latency: 10ms
CPU Load: 95.5%
Rare dropouts.

Buffersize: 1280-2048
Latency: 13ms
CPU Load: 92%
No dropouts

Frequency: 44.1KHz

No dropouts with this samplerate!

Buffersize: 64
Latency: 2ms
CPU Load: 52.2%

Buffersize: 128
Latecncy: 3ms
CPU Load: 48.2%

Buffersize: 256,384, 512 & 768
Latency: 6-47ms
CPU Load 46-42%

No improvement with larger buffers, even when going as far as 2048 buffersize I only managed to get to 42.3% load.

After a fresh install of everything, I’m suprised not to notice bigger drops in CPU load at higher buffersizes. I also noticed that Renoise’s CPU load meter only represents one core or 50% of the total, even though taskmanager’s performance graphs clearly show load being distributed amongst both cores. Setting renoise’s affinity to either single CPU caused CPU_LOAD_BENCH.rns to creep up to 90.2%, and did not rectify when setting affinity to both cores again.

Some interesting results, which makes me think I won’t need an upgrade for a while :)

36

Here are my crappy results. :lol:

Intel Pentium 2.4 GHz L2 On-Board Cache 512kB. 2x512 MB Kingston DDR-SDRAM, 333 MHz.  
ASUS P4PE motherboard, 533 Mhz FSB. Renoise 1.5.2 on Windows XP Professional.  
M-Audio Delta-44. All results at 44.1 kHz.  
  
- 2 ms 128 samples test.rns 84% CPU_LOAD_BENCH.rns 100%  
- 5 ms 256 samples test.rns 80% CPU_LOAD_BENCH.rns 100%  
- 11 ms 512 samples test.rns 75% CPU_LOAD_BENCH.rns 100%  
- 23 ms 1024 samples test.rns 72% CPU_LOAD_BENCH.rns 100%  
- 46 ms 2048 samples test.rns 71% CPU_LOAD_BENCH.rns 100%
37

A64 3200+ @ 2500 (~3950+), Corsair VS 2x512 (dc) @ 208 (2.5, 3, 3, 7, 1T), ASrock 939dual-sata2, SBlive! with KX drivers:

CPU_LOAD_BENCH.rns, 44.1khz:

23ms - 49.5%
11ms - 53%
5ms - 57.6%
2ms - 65.5% (with crackles)

38

Don’t get me wrong, but with my old P3-1200 I get dropouts, clicks and 99% with both of files at any sample rate and buffer size :)

Would be interesting to use render time instead peak(?) CPU usage, imho.

39

I have a Intel Core Duo T2400 1.83 GHz laptop I can test it on. I’ll do it when I get the time.

40

finally got rid of my p4 northwood after 2 1/2 years.
so time for some benches again ;)

CPU Core 2 Duo E6600 (Conroe) @3.6Ghz (400mhz FSB)
MOB Asus P5W-DH Deluxe, Intel 975x
RAM 2x1024MB DDR2 @800Mhz DC, CL4-4-4-5
GFX 7600GT, PCIe
SFX EMU 1212m, latest drivers

CPU_LOAD_BENCH.RNS
all tests run at 44.1 khz /w ASIO
[02ms] peaks at ~29%
[10ms] peaks at ~24%
[20ms] peaks at ~23%

comparing these values to my previous system (p4 3.0@3.8ghz), i almost trippled the performance.
damn i love my conny :D

@FuX
thanks for considering my request.
but time has passed and CPU evolution as well.
so the Pentium-M has after all lost my attention.
nevertheless, your results would be interesting of course.