New, Updated SonarTest

I built it myself.

ORIGINAL: Jim Wright

Woody, Zendin --

RE: Poor performance numbers with 845-based P4 motherboards.

Jim, I hadn't noticed this before, but I wanted to thank you for putting the time in to construct this very cogent, informative post. I appreciate your input very much. I'll have to research this a little. For now, I've switched to the EMM drivers and have a very stable, decent-performing system, but if I could spend $100 or so and have a really great system it may be worth it in the long run.

Thanks!
Mike

Where did you get your parts? And how do i go about guaranteeing that I get a matched pair of RAM. If the company I order from says it is, how will I know if it truly a matched pair? Thanks in advance.

regards,
John

ORIGINAL: JAB3

Where did you get your parts? And how do i go about guaranteeing that I get a matched pair of RAM. If the company I order from says it is, how will I know if it truly a matched pair? Thanks in advance.

regards,
John

I got most everything from PCClub (.com). The Corsair TwinX is a matched two sticks of Corsair XMS - hence the name Twin X. It's been super stable and I'm very happy so far. Left it running for two days solid now and testing as I had time.

ORIGINAL: zendin

ORIGINAL: Jim Wright

Woody, Zendin --

RE: Poor performance numbers with 845-based P4 motherboards.

Jim, I hadn't noticed this before, but I wanted to thank you for putting the time in to construct this very cogent, informative post. I appreciate your input very much. I'll have to research this a little. For now, I've switched to the EMM drivers and have a very stable, decent-performing system, but if I could spend $100 or so and have a really great system it may be worth it in the long run.

Thanks!
Mike

You're very welcome! Good luck, and please post your results if you decide to give it a try.

Scott -- any thoughts on whether moving up from an 845PE will make a noticeable difference?

- Jim

2-10-2004: Here's a link to a page giving memory performance results for various Intel chipsets: http://tech-report.com/reviews/2002q4/i845pe-ge/index.x?pg=5 The 845PE is actually better than the 850E with RDRAM (my setup) in terms of memory latency (rats!) but overall memory bandwidth is better with RDRAM (and of course with 865PE, 875P).

Okay, where'd them there DP Opteron 248 specs come from, eh?! Care to share the hardware list for this beastie? Please, please, please....

ORIGINAL: Resort Records

Okay, where'd them there DP Opteron 248 specs come from, eh?! Care to share the hardware list for this beastie? Please, please, please....

The 248 scores are from a system I am putting together for a customer.

Dual Opteron 248
Tyan K8W S2885
4GB Corsair XMS PC3200 memory (dual-channel per-CPU)
ASUS GeForceFX 5200 dual-DVI AGP video
PNY Quadro4 quad-DVI PCI video (total 6 heads)
HighPoint 1820 64bit/133MHz PCI 8-way SATA RAID controller
etc, etc...

:)

-S

Hi Scott

Some "new" figures for my o/clocked P4 2.4C rig.

Up unitl a few days ago, it was running at 3.0gig = 12x250mhz.

I have now optimized the o/c further and it is running stable at 3.2gig = 12x267fsb - memory is still at 1:1 with 2x512 in dual channel mod.

It's peformacve now would equal to a "3.4gig" given the fsb boost - noting of course that although the 3.4gig P4 dont exist yet, they will still only run at 200mhz fsb x 17 which I would suggest would not be quicker than 267mhz x 12 .

As before - I use ASIO drivers only and H/T and MP is turned off due to Kontakt "instabilities" with Sonar 3.1's MP engine [Kontakts problem not Sonars' ]

New figures for this " 3.4 " P4 are:-

46.4 24
23.2 26
11.6 31
5.8 39
2.9 57
1.5 92

Boosting the fsb to 267 from 250 has given something in the order of about an overall system boost of %7 - nothing to sneeze at - also, at 250fsb, 1.5ms would simply not even engage - now at 267fsb is still stable at %92.

amazing to think that a P4 has to hit a souped-up 3.4gig to just match the Athlon64 3200 down to 5.8ms - at 2.9 and 1.5, the Athlon 64 3200 is sitll %19 and %27 [respectivley] quicker than this souped-up 3.4gig - astounding really if you think about it - no wonder my next rig is going to be an Athlon64 3XXX.

Timboe

Maybe this should be a separate post. If so, apologies. I clicked on the Sonar3Test link and downloaded the zip file, which contained the .CWP file. What about the audio files? The original SonarTest was a bun file with the audio in it. Am I missing something?

ORIGINAL: kellyg

Maybe this should be a separate post. If so, apologies. I clicked on the Sonar3Test link and downloaded the zip file, which contained the .CWP file. What about the audio files? The original SonarTest was a bun file with the audio in it. Am I missing something?

This test is designed to stress CPU only... so all tracks are input monitored. There is no audio... and nothing to hear. Just note CPU usage at various latencies with the audio engine on.

-S

Thanks. I learn a lot from you guys.

The "delta" between an Athlon 64 system and an XP3200 (incl consideration of the mobo) is $300 or more. Maybe $400 if you go with the XP2500 which, according to other BB's I've read, runs nearly as well as the XP3200. I'm just starting out and don't consider that I will be recording more than 2-4 tracks simultaneously, nor do I envision projects with particularly high track counts. Most of my tracks will come from outside audio sources, but I do imagine that I'll use some soft synth sources as time goes on. Considering that most people seem to "trade up" systems every 2 years or so, would a good XP system be a logical starting place for me? There seems to be a lot to shake out over the next year or so in the transition to 64 bit.

bump

Here is my setup:

HP a410e
Athlon 3000+
1024 GB DDR RAM (2x512)
Western Digital 120 gig 7200 RPM 8mb cache
Western Digital 80 gig 7200 RPM 8 mb cache
NVidia GeForce 4 440MX
M-Audio Audiophile 2496
Windows XP Pro SP1
Sonar 3.1.1

WDM with M-Audio Delta driver .36
DMA 640

46.6ms 34%
23.2ms 36%
11.6ms 44%
5.8ms 45%
2.9ms 47%
1.5ms 49%

I'm actually quite shocked at how well this system performed. How is it that these numbers turned out so good. It never went over 50%!

Just got a new Athlon64 laptop, so I thought I'd post up.

eMachines M6805
Athlon64 3000+ Mobile Processor
512 GB DDR RAM PC2700
Fujitsu 60 gig 4200 RPM 2mb cache (built-in drive)
2 X Maxtor 160 gig 7200 RPM 8 mb cache via Oxford 911g Firewire enclosure
ATI Mobility Radeon 9600
M-Audio Firewire 410
Windows XP Pro SP1
Sonar 3.1.1

ASIO

At stock 1.8ghz (9x200):
46.6ms 31%
23.2ms 33%
11.6ms 37%
5.8ms 42%
2.9ms 53%
1.5ms 77%

OC'd to 2.0ghz (9x222 and extremely stable):
46.6ms 28%
23.2ms 30%
11.6ms 33%
5.8ms 39%
2.9ms 48%
1.5ms 68%

I'm thrilled with this setup. This laptop is a fantastic value at only $1300 from BestBuy (after $250 rebates). I finally have my mobile studio

JP

ORIGINAL: JoePaz

Just got a new Athlon64 laptop, so I thought I'd post up.

eMachines M6805

Hey, how's the build quality and noise level of that thing? Glad you're liking it.

I assume the hard drives didn't come with the system, right?

The build quality is excellent. For the most part it's totally silent. The fan kicks on once in awhile, but I have to put my ear to the machine to see if it's actually on. Its got a beautiful 15.4" widescreen display (1280x800), and it's a joy to work on. The only 2 negatives I have are the slow 4200rpm drive, and the memory upgradability. It comes with 2 256mb SODIMMS, only one if which is really user accessable. The second one is difficult to get to under the keyboard. I'd love to expand it to 1.25 gb, but 1gb SODIMMS are ridiculously expensive. I'll probably just get a 512mb SODIMM and go to 768mb.

This isn't the same eMachines company that made those crap machines in the late 90's. In fact, they've just been bought by Gateway. They'll continue the eMachines brand because of its relationships with retail outlets like BestBuy and Circuit City. You can check out some other opinions at http://notebookforums.com/forumdisplay.php?f=76 , http://www.hardforum.com/forumdisplay.php?s=&forumid=73 , and http://forums.amd.com/index.php?showforum=4 .

You can check the specs at http://www.emachines.com/products/products.html?prod=eMachines_M6805 . In addition to that $100 rebate from eMachines they have listed, BestBuy had an additional $150 rebate. I can't believe the amount of machine I got for $1300. And its kicks ass for gaming too.

Oh yeah, I already had the firewire drive setup. It didn't come with the machine.

JP

When you are posting latency setings; for example:
46.4ms........28%
23.2ms........33%
11.6ms........39%
5.8ms..........49-52%
2.9ms..........68-73%
1.5ms..........N/A

is this from the Buffer size Latency slider or Effective Latency at 44.1khz ?

Hi, I just bought an AMD64 3000 on a MSIk8t MOBO. I was expecting your results but my surprise has been that I have a 10% of more cpu used on each latencies. I wonder if the problem could be my memory setup because I use my old 764megs (266Mgz) and the seller from the shop told me that if I dont use a 400mgh memory then the cpu is not working properly.

So ,should I buy new memory and throug to the bin my old 764Mb(266mgh) for getting your results?

I use:
RME Hammerfall 9652
SATA 80gigs
764ddr (266Mhg)
AMD64 3000

(in the text my results are 10% of more cpu usage)

ORIGINAL: Danirustic


I use:
RME Hammerfall 9652
SATA 80gigs
764ddr (266Mhg)
AMD64 3000

(in the text my results are 10% of more cpu usage)

Danirustic,

Who's specs are you comparing yours to?

Oh sorry I see it was to Master Music. I'm kind of interested in your set up because it is similar to one I'm looking at in that it has a MSI mb.

Can anyone tell me what MSI motherboards are like? Its also got a Via chipset.

Lex

Finally figured out how to perform this test.

2.9----60-68%
5.8----45-48%
11.6---35-39%
23.2---29-33%
46.4---29-30%

ORIGINAL: Danirustic

Hi, I just bought an AMD64 3000 on a MSIk8t MOBO. I was expecting your results but my surprise has been that I have a 10% of more cpu used on each latencies. I wonder if the problem could be my memory setup because I use my old 764megs (266Mgz) and the seller from the shop told me that if I dont use a 400mgh memory then the cpu is not working properly.

That is very likely your problem. You are severely hindering the CPU's performance with that memory.

-S

Results are about the same as others...I don't think drives or video matter much in this test do they?

Sonar 3.1.1, Echo Gina24 w/WDM driver

Results:

46.4: 32-33%
23.2: 34-35%
11.6: 41-44%
5.8: 54-58%
2.9: 81-87%
1.5: Dropout

Athlon XP 2800
MSI K7N2-Delta L Motherboard
DDR 3200 set up as dual channel

I heard of this thing called Liquid DAW so I tried to make one and now it wont turn on... Was I supposed to use salt water instead of fresh?

Six weeks, seven pages, and still no Centrino numbers???

Sonar 3.11 , XP2500 @ 3200, NF2, Hoontech value24/96 running 8.1 drivers

ASIO

23.2 ms - 32%
11.6 ms - 39%
5.8 ms - 48%
2.9ms - 68%
1.5ms - dropout

WDM

23.2ms - 32%
11.6ms - 40%
5.8ms - 50%
2.9ms - 72%
1.5ms - dropout

Another data point:

Athlon64 3400
Gigabyte K8N Pro
1gb (2x512) Corsair TwinX 3200 DDR
Lynx 2 in ASIO
Sonar 3.1.1
Win XP


23.2 - 26%
11.6 - 29%
5.8 - 32%
2.9 - 41%
1.5 - 58%
.7 - 93% (although the engine ran you can't really do anything at this rate or the engine will dropout)


Scott

Late, I know. Not sure if I did this right as I was never offered the 1.9 or whatever but here goes.

MOBO MSI K7N2 Delta-ILSR
CPU Athlon Xp Barton 2500+ OC'd to XP 3000+ 2.10 GHZ
512 MB PC3500 Crucial. (ONE STICK)
Audiophile 2496 M-Audio WDM .36 drivers


46.4 ---- 31%
23.2-----33%
11.6------38%
5.8--------47%
2.9--------67%
1.x-------never gave me the option

I imagine those numbers would improve when I get around to OC'ing to XP3200+ and added another stick of RAM in order to use the "dual" function DDR and my MOBO are designed for.

Wheee....

Athlon XP 2500+
Epox 8RDA+
256 MB RAM
Windows XP Pro SP1
M-audio Delta 66 (v.27)
Sonar Producer 3.1.1

Latency ASIO WDM MME
-----------------------------------
46.4ms 35% 36% 40%
23.2ms 40% 40% 43%
11.6ms 47% 48% 53%
5.8ms 59% 61% 69%
2.9ms 85% n/a n/a
1.5ms n/a n/a n/a


Thanks for the test Scott.

wren

Wow -

My P4 system really gobbles the CPU...

Dell 4550
P4 2 ghz
640 MB PC2100 Ram
Windows XP - Home
7200 RPM 60 gig system drive
7200 RPM 80 gig storage drive


Roland SI-24 Surface with RPC-1 Card

All devices and background items disabled.

ASIO Drivers

2.9ms - nope
5.8ms - 75%
8.7ms - 64%
11.6ms-56%
17.4ms-49%
23.2ms-47%
46.4ms-42%

Why is my system so CPU Hungry?

Hmmmmmmmmmmmmm????????????

Arman,

As far as I can Google, the Dell 4550 uses an 845-series system chipset.
The more recent 4550's use an 845PE, I'm not sure what your system uses.
It should say in the 'Technical Specifications' section of the users' guide.

The original 845 was a real dog as far as memory performance goes. The 845PE is better, but still no match for more recent 865 or 875 system chipsets. Other Sonar users with 845xx Intel systems have reported seriously lacking performance -- search this thread for past messages ('845' should do it).

Memory performance appears to make a big difference as far as SonarTest latency is concerned. The Athlon64's seem to do so well (at least in part) because they have a memory controller integrated on the same chip as the CPU. With an Intel system (and older AMD systems), the memory controller is on a separate chip (part of the 'system chip set'), and acts as a traffic cop. With an 845, it acts more like a cranky traffic cop on a really muggy August day in NYC in the middle of rush hour....

- Jim

Hi Scott !

Just did a relatively minor CPU upgrade to my DAW rig - went from an O/C 2.4 P4 to an O/C 2.8 P4 - results are as follows:-

- Asus P4P800
- 2 x 512 DDR 500 @ 1:1 ratio
- Intel P4 2.8C [800fsb - *not* Prescott] O/C stable to 3.3 gig [ 14 x 236fsb ]
- H/T OFF in the BIOS
- MP OFF in Sonar
- Delta 66 Card
- ASIO Driver
- Delta .36 release
- WinXP Pro
- Sonar 3.1.1

1.5ms - NO GO - mid-high 90's but dropping out

2.9ms - 55%

5.8ms - 37%

11.6ms -27%

23.2ms - 22%

46.4ms - 19%


Timboe

This is a great thread, very informative. As a new member of the board, a warm hello to everyone.

I posted elsewhere that I just bought Sonar PE 3, and now I need something to run it on. (I also have a couple of guitars, a Yamaha A3000 sampler and RS7000 sequencer, and a MicroKorg synthesizer.)

My question is this: Should I buy/ build a DAW based on the AMD 64 3200, or should I go with a Dual Opteron 242 set-up? This thread points me to the
AMD 64 3000, 3200 and 3400 CPUs.

BUT, there is also some info pointing in the direction of the dual opteron system, but not as much. Not as much discussion or analysis of the dual opterons. I can't afford much more than two 242's right now. (Dual 248's would be a no-brainer, but very expensive.)

I was all set to get an AMD 64 3200 CPU system, but now I am wondering about a dual opteron 242 system. Couldn't I upgrade the 242's to 248's later on as the 248's come down in price? Would the 242's be much slower than the 3200? Do the 242's give me more room to expand waht I am doing?

I am having a hard time finding info comparing opterons to the 64's for DAW performance. AMD's website seems to push you toward the Opteron for digital audio, but then they mainly talk about animation when you dig deeper. I've looked at ANANDTECH and SLASDHDOT and INQUIRER and still can't find what I'm looking for. If any of you can offer advice, please do.

Many thanks!!

Thanks for your input!

Well, seeing as how this thread is still going on strong, I guess I should post the results with my new rig:
AMD 64 3000+
MSI K8T-800 NEO FIS2R (BIOS 1.20)
1GB PC3200 (400) Crucial RAM (2x512)
Sapphire ATI 9600xt Ultimate
IBM Deskstar 30GB (OS & APPS)
Maxtor SATA 120GB (Audio Files)
Windows 2000 Service Pack 4
Frontier Labs WaveCenter PCI (3.01 drivers)

WDM:
46.4ms........31%
23.2ms........32-33%
11.6ms........36-38%
5.8ms.........41-42%
2.9ms.........56-58%
1.5ms.........87-90%

Here's mine,

ASUS A7N8X-Deluxe MB
AMD XP 2600+ no O/C
Corsair 400mhz RAM 2x512 (Setup for dual channel memory config)
Matrox AGP450 dual head Video
Maxtor 40gb 8MB cache boot drive
Maxtor 80gb 8MB cache data drive
Aardvark Q10 audio interface with 7.13 drivers
WinXP Sp1
Sonar Prod 3.1.1

SonarTest results are as follows;

Latency ASIO A/WDM(aardvarl's proprietory driver)

46.4 33% 38-40%

23.2 35-36% 49-50%

11.6 42% 67-69%

5.8 54-55% In red Warning! But audio engine still active

2.9 76% Won't run at all

1.5 N/A N/A

Thanks for going through the trouble Scott,

All the best,

Wayne

System:
Mobo: ECSK7S5A
RAM: 512Mb PC 2100 Muskin Hi Pref at cas 2
Processor: Athlon XP 2400 266 FSB

M-Audio Audiophile at 128 samples(WDM/Asio made no difference,so I stayed at WDM)


46.4ms: 38%peak
23.2ms: 37%peak
11.6ms: 45%peak
5.8ms: 60% peak
2.9ms: 90+warning!
1.5ms(at 64 samples): no
After 23.2 ms,it stayed at around 38% as i wen't higher.I'll bench my other 2 machines later(2600/333 and 2800/333fsb).

Hmmm, given the others' results, I'd expect better numbers than this...but I have had a problem lately in the middle of mixing where the CPU wacks out to 100% and stay there a ten seconds before returning, then requiring restarting SONAR to get it to run again...

Anyways, here goes:

Intel P4 2.4GHz
no clue on the Motherboard
1GB PC2100 DDR (2x512MB)
Maxtor 120GB 7200rpm drive
WindowsXP Professional SP1
Maudio Audiophile 2496

WDM

46.4ms........42%
23.2ms........46%
11.6ms........53%
slider will not go below 11.6ms...?

bob.

My results
Intel Centrino 1.4 GHz laptop
Motu 828 MKII
41% 46.4ms
46% 23.2ms
61% 11.6ms
65% 5.8ms
78% 2.9ms
refuses 1.5

slider will not go below 11.6ms...?

bob.

You need to configure the Audiophile's software control panel for a smaller buffer.

-S

Scott,

Are you sure that the numbers for the A64 3200+ are correct on your graph? On the graph, it looks like at 5.8ms the A64 3200+ is at about 34%. I can only find two people in this thread that posted info on A64 3200+ systems, and they both state around 42% at 5.8ms.

I am just trying to determine how much extra hp I would get with the A64 vs. my current XP 2600+.

Thanks,
Tom Boughner

I have been trying to get the same answer/advice. Please let me know if anyone advises you.

Thanks,

Someone asked about the 3200+ in another thread as well...

I used a system I had just completed for someone to get those numbers. In every case, I've used the best results available for any given CPU. In a few cases, those results did not come directly from this thread.

-S

Thanks for the clarification, Scott.

In that case, would you mind sharing what mobo you used?

Thanks,

It was Gigabyte's K8N Pro. I also used low latency Corsair memory. Memory latency does seem to impact this test when it comes to A64 CPUs.

-S

Thanks Scott - that's good to know.

Tom

Here's mine:

P4 2.8e (1MB L2, 90nm)
Abit IC7-g
1GB (512MBx2) PC3200 (non-ECC) dual chan mode
Nvidea MX440 64MB DDR
Segate 80GB SATA audio
WD 80GB ATA100 system
Segate 20GB ATA100 SWAP
Asus 52x CD-Rom
CyberDrive CD-R,RW
Antec true 430W PS
Windows XP Pro SP1a
M-Audio Delta 1010

ASIO---------MP----------SP

46.4ms........28%--------32%
23.2ms........30%--------34%
11.6ms........34%--------39%
5.8ms..........44%--------48%
2.9ms..........61%--------66%
1.5ms..........N/A---------N/A

WDM---------MP----------SP

46.4ms........29%--------34%
23.2ms........31%--------36%
11.6ms........36%--------41%
5.8ms..........N/A--------N/A
2.9ms..........N/A--------N/A
1.5ms..........N/A--------N/A

MP-for multiprocessor engine engaged
SP-for multiprocessor engine off

Intersesting thread, good stuff! Now I'm back to ASIO for the 1010!

Processor AMD Athlon(TM) 64 3200+ 2.00 GHz
Operating System Microsoft(R) Windows(R) XP Professional
Memory 1.0GB DDR SDRAM
Hard Drive 60 GB 7200 RPM Hard Drive was factory 4200
Primary CD/DVD Drive 4X DVD+RW/R & CD-RW Combo Drive
Communication Built in 56K Modem+10/100 LAN & 54g(TM) Integ. Broadcom 802.11b/g WLAN & Bluetooth
Display 15.4" WVA WXGA (1280x800)
Graphics Card 64MB NVIDIA(R) GeForce(TM) 4 440 Go
Primary Battery 12 Cell Lithium Ion Battery
Tascam FW-1884

ASIO
46.4ms........34%
23.2ms........37%
11.6ms........43%
5.8ms.........50%
2.9ms.........62%
1.5ms.........75%

Forgot to mention this is a LAPTOP

Processor AMD Athlon(TM) 64 3200+ 2.00 GHz
Operating System Microsoft(R) Windows(R) XP Professional
Memory 1.0GB DDR SDRAM
Hard Drive 60 GB 7200 RPM Hard Drive was factory 4200
Primary CD/DVD Drive 4X DVD+RW/R & CD-RW Combo Drive
Communication Built in 56K Modem+10/100 LAN & 54g(TM) Integ. Broadcom 802.11b/g WLAN & Bluetooth
Display 15.4" WVA WXGA (1280x800)
Graphics Card 64MB NVIDIA(R) GeForce(TM) 4 440 Go
Primary Battery 12 Cell Lithium Ion Battery
Tascam FW-1884

ASIO
46.4ms........34%
23.2ms........37%
11.6ms........43%
5.8ms.........50%
2.9ms.........62%
1.5ms.........75%

My System:

MB: Albatron PX845PE Pro
P4 2.4, 533 FSB
512mb PC2700 DDR
Seagate 80gig 7200 rpm
WD 80 gig 7200 rpm
Maxtor 30gig 7200 rpm
GEForce MX 440 64mb dual head video
Samsung CDR drive
RME HDSP 9652 using ASIO drivers


The Numbers:

1.5 = nope
2.9 = 85%
5.8 = 59%
11.6 = 45%
23.2 = 38%
46.4 = 34%

ORIGINAL: Stratman

Running a P3 Dual 1.4ghz Intel Tualatin CPU's.
Front Buss is 133
512mb RAM PC133
1 - WD8mb ATA 100 40gig Hard drive for OS
1 - WD8mb ATA 100 80gig Hard drive for Data

On 2000 PRO and S3.1

With MP mode
46.4 = 46%
23.2 = 51%
11.6 = 57%
5.8 =70%
2.9 =Redline Warning ( No Dropout)

Without MP mode
46.4 = 65%
23.2 = 70%
11.6 = 80%
5.8 = Redline Warning ( No Dropout )
2.9 = Redline Warning ( No Dropout )

With WDM drivers on a Delta Omni .29 drivers
I could not get down to 1.5 with my drivers and soundcard .

My system rarely drops out when redlining .It seems to roll over to the second cpu . Interesting . I thought I had a smokin pc till I saw some other stats .

Thanks Scott

Paul

I have a SINGLE processor the same as above and I'm slightly better than stratman(in MP mode) excpt when he gets down to 5.8 at 70 ms and I'm at 75ms. This is with me OCed to 1500. One odd thing about this test is when I OC to 1660 there is no difference, whereas on other tests, the improvement is 9-10%. Porque? (why in Spanish).

Just speaking hypothetically.....what if I overclocked my Athlon XP 3000 to (3000 mhz) and say it was equivalent to a athlon xp 4000
(if it were rated there) how would it stand up to the others? like the 64's opteron ..( ijust checked the price on most of those and they are expensive...even justifiying a cooling system....)
here are some listing of what is possible with my setup with an abit NF7-S

www.overclockers.com

or does anyone have a system like that were they can submit for review........just a thought

P.s. im not going to overclock for a while........maybe after a year and a half...until i do more research
on stability, longevity of chip, perfomance!

As there haven't been many Centrino data points, and
none above 1.5GHz, here's a full set of numbers for my
newest machine.

IBM Thinkpad T41
"Centrino" Pentium M @ 1.7GHz
1GB PC2700 DDR SDRAM
Windows XP Pro
Sonar 3.1.1
RME HDSP Driver v2.70 (HDSP PCMCIA/Multiface)

Latency - Reported CPU Load
1.5ms - 91% (+/- 1)
2.9ms - 65% (+/- 4)
5.8ms - 47%
11.6ms - 40%
23.2ms - 36%
46.4ms - 32%
92.9ms - 30%
185.8ms - 30%

The curve is actually somewhat distinct from the other CPUs
plotted on Scott's graph, and I think it ought to be plotted there.

This machine actually runs some Reaktor ensembles fractionally
faster than my 2GHz Opteron system. The Centrino's a real "sleeper".

ORIGINAL: KevinK

The curve is actually somewhat distinct from the other CPUs
plotted on Scott's graph, and I think it ought to be plotted there.

Well, of course it ought to be plotted.

I'll be able to get to the file in the next couple days. It looks to be generally on par with a 2.6-2.8GHz P4, with the advantage of being able to run at 1.5ms.

-S

when will P5 be available ? (and what are the "suggested" specs)

mb

ORIGINAL: SAMPLIFIER
My question is this: Should I buy/ build a DAW based on the AMD 64 3200, or should I go with a Dual Opteron 242 set-up? This thread points me to the
AMD 64 3000, 3200 and 3400 CPUs.

BUT, there is also some info pointing in the direction of the dual opteron system, but not as much. Not as much discussion or analysis of the dual opterons. I can't afford much more than two 242's right now. (Dual 248's would be a no-brainer, but very expensive.)

I'm pretty sure that Scott's chart and this thread is the best information available anywhere on how well Sonar actually works on different CPUs. Bear in mind that the stuff on the chart reflects the FASTEST performance anyone has reported from a given CPU, so when you look at the numbers, you're USUALLY seeing a comparison of well-optimized DAW systems. An off-the-shelf dell or Gateway with a CPU that scores well on the chart may be easily outperformed by a highly-optimized dedicated DAW machine with a lower-ranking CPU. So make your decision based on your WHOLE budget. I don't think I'm going out on a limb by saying that MOBO, hard disk, and memory decisions can have as much effect on DAW performance as CPU can. In addition, if you're a bedroom producer who must track and engineer in the same room, having a quiet machine will have as much effect on the quality of your recordings as the computing power will. Audio recording is something that can really push a computater to it's limits, and the weakest link is always the most important one. So draw up a budget, prowl around the forums, and pick out a MOBO, chipset, RAM brand, and disk drive setup, and then buy the best-performing CPU you can afford with whatever's left. That's my not-very-expert advice, anyway. Cheers, and have fun.

The chart has been updated with Centrino results...

-S

Where is the chart?

Thanks.

The very first post in this thread.

-S

Thanks. After checking liquiddaw.com, cakewalknet.com, I tried but didn't wait long enough for the first page of this thread to load. The joys of dial up.

Thanks again.