SFP and AMD X2 ?

Shayne White · Post by **Shayne White** » Thu Dec 08, 2005 10:07 pm

OK, I found the page where you got the hint from...I tried both /intaffinity (setting interrupts to one cpu) and /onecpu (turning off multiprocessing) -- not at the same time, of course. To my surprise, it sill crashed!! It has nothing to do with the CPU. The only other explanation is the fact that in ACPI multiprocessing mode, it's using more than 16 IRQs...in ACPI singleprocessing, it's not using more than 16. I tried using /pcilock, but that didn't chnage anything. Well, I'm glad my CPU isn't defective, but I don't know how to make this work. Any ideas? My motherboard doesn't have "force an IRQ number" options, and Windows refuses to follow the BIOS IRQ settings in multiprocessing mode.

Shayne

fra77x · Post by **fra77x** » Fri Dec 09, 2005 9:12 am

I can't also change the Ulli without an instant freeze, but i can work with it, as i just change the ULLI after SFP restart. Except that my system works without any problem.

jea · Post by **jea** » Fri Dec 09, 2005 11:16 am

Hi,

Shayne, did you use GigaByte K8NS Ultra-939?

You can specify the irq's in bios if so.

/INTAFFINITY in boot.ini
/SETAFFINITY for sfp to the highest cpu nr.

Shayne White · Post by **Shayne White** » Fri Dec 09, 2005 11:49 am

I use the Gigabyte K8NSC-939, and there's no option to change IRQs in the BIOS. Windows refuses to follow the motherboard's IRQs even when /pcilock is in boot.ini.

/intaffinity didn't help.

The only thing that makes it work is changing Windows's Computer driver to ACPI without multiprocessing, and then the IRQs change to 16 only, and Scope works perfectly.

As I said in a previous message, when I was using ACPI multiprocessing with 24 IRQs, using Sonar with ASIO made the ASIO driver lock up after a few minutes, which goes along with changing the ULLI setting.

The issue is obviously not multiprocessing itself, it's the way the IRQs are distributed in multiprocessing mode. I don't know how to fix that.

Shayne

symbiote · Post by **symbiote** » Fri Dec 09, 2005 2:00 pm

Have you tried hitting Ctrl-F1 in the BIOS to get the advanced menu?

jea · Post by **jea** » Fri Dec 09, 2005 4:49 pm

Hi, I think that

/INTAFFINITY
Directs the standard x86 multiprocessor HAL (Halmps.dll) to set interrupt affinities such that only the highest numbered processor will receive interrupts. Without the switch, the HAL defaults to its normal behavior of letting all processors receive interrupts.

and the

/PCILOCK
Stops Windows from dynamically assigning IO/IRQ resources to PCI devices and leaves the devices configured by the BIOS. See Microsoft Knowledge Base article Q148501 for more information.

switches should work, you might have to install *not in acpi mps. but in mps*....

Here is a simple search on google:

http://www.google.no/search?hl=no&q=Hal ... odes&meta=

I think there is more to this, but shouldn't CWA be aware of such things? I mean if CWA really knows the Winplatform inside/out as a developer for that platform?

Then all the whining and moaning would end, and we could make music without thinking too much about all this shit!

astroman · Post by **astroman** » Fri Dec 09, 2005 11:00 pm

All the whining and moaning seems to refer to not even a handful of people, but according to

On 2005-11-03 16:54, buyakasha wrote:
Hey claudioDupgraded my . I've system about a month ago which uses AMD X2 with SFP and I can tell you that it works damn good.
...
I use 2 Scope pro cards (max 30 (28) DSP's plus a UAD card. My motherboard is the Gigabyte K8NS Ultra 939 using the nf3 chipset. This board has 5 PCI slots and an AGP graphics slot and supports the AMD64 dual core.

the problem is not Scope specific.
I completely agree with you that it's all sh*t, an inherent feature of the Windows platform in general.
Not that I second religious viewpoints in the 'this and that is better...' style - it's just what experience tells.
One often has to live with it, tho.
my respect for Shayne's endurance

cheers, Tom

[ This Message was edited by: astroman on 2005-12-09 23:01 ]

jea · Post by **jea** » Sat Dec 10, 2005 7:20 am

Well, Mr Astroman, it is Scope specific, or should I assume CWA is using PII, PIII and old mobos?

Ahem ......

The world is moving on, and this CWA should do also, providing a solution, to make things work, without vasting time, and if you look back, look at promises in old brouchures.

That said, I am surprised that the driver is not updated since 2002.

May I suggest that the only solution to why, perhaps is that CWA might not have any qualified persons left to make new drivers, working under dual cpu's etc.

Word of mouth is responsible for people not buying into it/ not choosing it, or choosing it and then leaving it, hoping for new drivers to arrive, choosing another solution.

Dual CPU was possible for NT4 -> ..... so it is not a new technology, we know that ....

It's not what we can do for CWA, it's what CWA can do for us. PERIOD.

I'm not whining, this is the fact.

So please CWA, release a driver that is compatible with dual core/cpu/hyperthreading, and who knows what!

After all, since it is the dsp's responsibility is to alter the sound, that shouldn't so hard?

I see most of the software running on the host (SFP/XTC) as merely a way to control the modules loaded into the dsp's.

A remotecontrol.

Is it that hard to make the communication between the remotecontrol (SFP/XTC), and the dsp's (hardware via driver ...) work?

Or is it back to Windows 98, CUSL-2, and PIII? And just use ADAT i/o? And the midi ports thats so fragile when we say BOOO! in the form of timecode, clocks, etc that they crash the entire machine?

It's a lot of if/then/what/etc to get Scope running smooth.

And just miss the Vinco, SPL stuff, etc as XTC plugs?

On 2005-12-09 23:00, astroman wrote:
All the whining and moaning seems to refer to not even a handful of people, but according to
On 2005-11-03 16:54, buyakasha wrote:
Hey claudioDupgraded my . I've system about a month ago which uses AMD X2 with SFP and I can tell you that it works damn good.
...
I use 2 Scope pro cards (max 30 (28) DSP's plus a UAD card. My motherboard is the Gigabyte K8NS Ultra 939 using the nf3 chipset. This board has 5 PCI slots and an AGP graphics slot and supports the AMD64 dual core.
the problem is not Scope specific.
I completely agree with you that it's all sh*t, an inherent feature of the Windows platform in general.
Not that I second religious viewpoints in the 'this and that is better...' style - it's just what experience tells.
One often has to live with it, tho.
my respect for Shayne's endurance

cheers, Tom

[ This Message was edited by: astroman on 2005-12-09 23:01 ]

astroman · Post by **astroman** » Sat Dec 10, 2005 9:38 am

Sir, we may disagree about which flavour of an OS deserves the attribute 'progress'...

but there is no doubt about the fact that Mr. Buyakasha got the thing running that you're discussing here

...unless you assume he's not telling the truth

if it would be a Scope problem, it would fail on him, too.

that a 7 years old Pulsar One doesn't integrate perfectly in the latest (almost) experimental chipsets isn't surprising at all. It's ridiculous to demand a manufacturer has to support all and every legacy device for all times - even companies with much larger resources than CWA don't do that.

cheers, Tom

jea · Post by **jea** » Sat Dec 10, 2005 5:23 pm

On 2005-12-10 09:38, astroman wrote:
Sir, we may disagree about which flavour of an OS deserves the attribute 'progress'...

but there is no doubt about the fact that Mr. Buyakasha got the thing running that you're discussing here
...unless you assume he's not telling the truth

if it would be a Scope problem, it would fail on him, too.

that a 7 years old Pulsar One doesn't integrate perfectly in the latest (almost) experimental chipsets isn't surprising at all. It's ridiculous to demand a manufacturer has to support all and every legacy device for all times - even companies with much larger resources than CWA don't do that.

cheers, Tom

Hi Astroman,

Mr. Buyakasha has not replied to my post in that thread, could be that he have not seen it, or maybe it doesn't work. The post does not indicate what mode he runs the dualcore in.

And one more thing, it's generally a Scope thing.

Might I also say, I have not even buildt it yet, but the parts are here, for 3 machines.

About, "that a 7 years old Pulsar One doesn't integrate perfectly in the latest (almost) experimental chipsets", we are not talking about a Pulsa One, and remember the chipsets were as "experimental" then as now.

What about new buyers today?

CWA doesn't tell people what it works nor works with, from ad's it is not mentioned at all....

With all respect, you jumped over my real critic of CWA, didn't you?

"It's ridiculous to demand a manufacturer has to support all and every legacy device for all times - even companies with much larger resources than CWA don't do that."

Read all the old brouchures for the SFP platform.

No, it's not ridiculous, as a pci soundcard from 5 years ago probably works today in todays computers, and we are talking about a product that CWA is selling today, yet everybody have problems with it, and have to make their computers less powerfull to make it work.

And about "the resources CWA have" (or don't have);

Since most people have problems with dual cpu (from some years ago's posts), with all respect, CWA why have you not fixed the drivers and the bugs?

Now this is RIDICILOUS!

This is like selling a car with defective steering and brakes, knowing it will crash! (as our computers eh, sometimes do ...)

eh?

symbiote · Post by **symbiote** » Sat Dec 10, 2005 7:16 pm

I have a fairly modern system (ADM64 3000+) and it works fine with CW cards. Dual CPU/core has always been a bit problematic over the years, if you overlooked this when buying a new machine it is entirely your responsability.

Why aren't you asking Microsoft why a driver that works in single core doesn't work with dual core? Isn't it their job to make multiprocessor/cores totally transparent in their OS?

There's a ton of posts on this forum about which systems work well with CW cards, so you really have no pretext.

jea · Post by **jea** » Sun Dec 11, 2005 12:30 am

Hi Symbiote,

dual cpu's have been there since NT4 for consumers. And this is before 2000. Hardware made before 2000 often work well in modern computers too, with or without dual cpu.

I assure you, my dealer told me that Pulsars were "futureproof", and was leading the most leading technology at that time.

And there were some other DSP choices at that time.

I choose Pulsar One.

But the time went, and I bought several other Creamware cards.

I now have a total of 7 cards, including a Creamware TripleDAT ISA.

But computers have been a lot more powerful also, and with the power we have today, sequencers get more powerfull too. But if we can't use our cards together with integration, we are kind of stuck here in single cpu land. I think that if Creamware (to live up to it's name) have to do this now. They should have done it several years ago, to protect their technology cashwise. Imagine what happened if XTC were working, dual cpu, and lots of new plugins were released in 2000-2003.... UAD and TC would have a hard competition, don't you think? And it must be easy to install, like UAD.

Since CW have not dealt with this issue, I am not sure they ever will, I don't think they are capable of it anymore, I don't think they have any qualified programmers to do it.

Otherwise they would have done it in 2000. To protect their business.

On 2005-12-10 19:16, symbiote wrote:
I have a fairly modern system (ADM64 3000+) and it works fine with CW cards. Dual CPU/core has always been a bit problematic over the years, if you overlooked this when buying a new machine it is entirely your responsability.

Why aren't you asking Microsoft why a driver that works in single core doesn't work with dual core? Isn't it their job to make multiprocessor/cores totally transparent in their OS?

There's a ton of posts on this forum about which systems work well with CW cards, so you really have no pretext.

astroman · Post by **astroman** » Sun Dec 11, 2005 2:55 am

On 2005-12-10 17:23, jea wrote:
...
About, "that a 7 years old Pulsar One doesn't integrate perfectly in the latest (almost) experimental chipsets", we are not talking about a Pulsa One, ...
...

With all respect, you jumped over my real critic of CWA, didn't you?

...
This is like selling a car with defective steering and brakes, knowing it will crash! (as our computers eh, sometimes do ...)
eh?

points 1,2 and 3

1 - Shayne has 3 cards, with 1 Pulsar one among them. I bet it's that card that 'disturbs' the setup. According to AndreD a Pulsar one doesn't even run on the latest Intel boards - this is about an AMD setup, but extrapolating may apply.

2 - No, I'd never jump blindly on critics.
You just generalize a bit too much from your own viewpoint and place it as a synonym for what the majority of customers (current and future) may expect.

3 - check out [link removed] about the truth in your comparison

cheers, Tom

ps that was supposed to be linked http://soli.inav.net/~catalyst/Humor/gm.htm

[ This Message was edited by: astroman on 2005-12-11 06:32 ]

astroman · Post by **astroman** » Sun Dec 11, 2005 3:20 am

On 2005-12-11 00:30, jea wrote:
... They should have done it several years ago, to protect their technology cashwise. Imagine what happened if XTC were working, dual cpu, and lots of new plugins were released in 2000-2003.... UAD and TC would have a hard competition, ...

this assumes that the technology you refer to IS their cash cow - it has been stated several times by (the old) CW that they ARE still in business due to several OEM deals and their broadcast solution in particular

Your statement would in fact be correct, if they had succeeded taking over (at least a part of) the ProTools market - but this didn't happen, so they (and we) have to live with what's left.

Honestly, I don't think we're doing bad this way. Even YOU have been happy last year with your system

It's still the best bang for THAT amount of cash, no matter on what it runs.

I dunno for what you start a run after the latest and greatest 'technology' (it's not my business anyway), but if you're in this as long as you pretend, a more 'realistic' estimation of the true meaning of buzzwords may apply

UAD and TC sell on big names and their respective image.
Whatever is released on SFP will never make it unless there's a BIG fat sticker on it. People do not by on facts, let alone their ears.

You may have read all those enthusiastic comments about B2003 and Profit recently - those devices were in the shop for years and they have not been particularly expensive...
So why didn't those folks buy them earlier ?
You DO know how many reverbs Warp69 has sold

All this doesn't happen due to a lack of support of a certain hardware, OS or driver technology, but because it's a non-pro market segment with a few (smart) dudes who take advantage of outstanding tools.
Would you place your investment there ?

cheers, Tom

jea · Post by **jea** » Sun Dec 11, 2005 4:53 am

Listen Tom,

there are various reasons to why I do things.

I can't tell you why.

I am not sure why you directed me to:

http://www.lisashea.com/index.html

But it sure is a nice site.

Perhaps you could explain for me.

I did not have all of my system up & go in the way I wanted it to, but it was workable.

I do not enjoy problems, but somehow, I try to solve them.

After a while you get tired of what suppose to be the "source" of the problems.

In the end, all it comes down to; "All I wanted to do was make music".

And I wanted it on one pc. Then I understood I cannot because of incompatibility.

This happened on a bunch of hardware.

You have seen the hardware I have, all I have bought.

I have not used it much. It does not work good enough.

astroman · Post by **astroman** » Sun Dec 11, 2005 6:19 am

you're right - they redirect

http://soli.inav.net/~catalyst/Humor/gm.htm
should do it

cheers, Tom

[ This Message was edited by: astroman on 2005-12-11 06:30 ]

symbiote · Post by **symbiote** » Sun Dec 11, 2005 8:02 am

You may be right, but like I mentionned, if you are at 7 cards total, you KNEW they didn't work well with dual core/dual CPU, so you still have no pretext.

Also, since CW cards aren't dumb I/O-less plugin accelerators, you could very easily put the cards in a single CPU computer, and hook it up to your dualcore CPU with ADAT. If you can afford a dualcore systeme, it shouldn't be too much of a problem to afford an additional ADAT interface to it also. Then you can just put the CW cards in a system that is known to be stable right away, and still get the benefits of dualcore. So it's not like you are stuck and can't do anything.

XTC will never work well, it's not a matter of throwing more programmers at the problem, it's a matter of trying to use a protocol made for software synths to integrate hardware devices, which is a flat-out braindead stupid idea, but required by hordes of wannabe-musicians who "just want to make music and couldn't possibly be bothered by technical stuff" (sounds extremely funny to me given that they all use modern sequencers, which are really extremely technical.)

The Virus TI, which promised "total integration" with the sequencer with a VST gimmick, was released almost 11 months late, and for 2 months the mailing lists were FULL, FULL of issues highly similar to the XTC ones. Worked for some people, not for others, works but clicks n pops, works but plugin-delay compensation/latency is unusable, crashes but managed to make it work after a full XP re-install (mm deja-vu =P), OMG it took Access 10 months to get this out the door why doesn't it work well for everyone OMG I'm selling mine and never touching Access products again, etc etc.

So if you want to have some XTC-mode that works, take your issue to Steinberg (who designed the VST protocol), take your issue to sequencer vendors and developers to try and get them to agree on a standard to integrate hardware (they're in competition technically, so good luck!!!), or even to Digidesign, if you bug them they might open up the whole TDM standard, just for you =P But before everyone agrees on a standard, or before Creamware releases their own sequencing package, you can totally forget about it, and there's very little/none Creamware or anyone else can do about it.

I might be wrong on this, but I don't think XTC, dual cpu support, nor more plugins would have helped Creamware at all. What *would* have helped them is to make all plugins crackable (appeals to the huge but very low fund cheapass VST weenies), OR have some brand sticker involving Waves or whoever else all over the place AND include a ProTools-comparable sequencer to solve the whole integration thinger. And charge more!

Finally, I don't know where you got the idea that most hardware made pre-2000 still works fine on modern computers. Most of that hardware *doesn't even have drivers for Windows XP*. But yeah, if you run Windows ME on that new dualcore, it'll work fine

PS Don't listen to dealers, next time contact the company itself =P

On 2005-12-11 00:30, jea wrote:
Hi Symbiote,

dual cpu's have been there since NT4 for consumers. And this is before 2000. Hardware made before 2000 often work well in modern computers too, with or without dual cpu.

I assure you, my dealer told me that Pulsars were "futureproof", and was leading the most leading technology at that time.

And there were some other DSP choices at that time.

I choose Pulsar One.

But the time went, and I bought several other Creamware cards.

I now have a total of 7 cards, including a Creamware TripleDAT ISA.

But computers have been a lot more powerful also, and with the power we have today, sequencers get more powerfull too. But if we can't use our cards together with integration, we are kind of stuck here in single cpu land. I think that if Creamware (to live up to it's name) have to do this now. They should have done it several years ago, to protect their technology cashwise. Imagine what happened if XTC were working, dual cpu, and lots of new plugins were released in 2000-2003.... UAD and TC would have a hard competition, don't you think? And it must be easy to install, like UAD.

Since CW have not dealt with this issue, I am not sure they ever will, I don't think they are capable of it anymore, I don't think they have any qualified programmers to do it.

Otherwise they would have done it in 2000. To protect their business.

jea · Post by **jea** » Sun Dec 11, 2005 3:10 pm

Symbiote,

I have no words.

JEA

On 2005-12-11 08:02, symbiote wrote:
You may be right, but like I mentionned, if you are at 7 cards total, you KNEW they didn't work well with dual core/dual CPU, so you still have no pretext.

Also, since CW cards aren't dumb I/O-less plugin accelerators, you could very easily put the cards in a single CPU computer, and hook it up to your dualcore CPU with ADAT. If you can afford a dualcore systeme, it shouldn't be too much of a problem to afford an additional ADAT interface to it also. Then you can just put the CW cards in a system that is known to be stable right away, and still get the benefits of dualcore. So it's not like you are stuck and can't do anything.

XTC will never work well, it's not a matter of throwing more programmers at the problem, it's a matter of trying to use a protocol made for software synths to integrate hardware devices, which is a flat-out braindead stupid idea, but required by hordes of wannabe-musicians who "just want to make music and couldn't possibly be bothered by technical stuff" (sounds extremely funny to me given that they all use modern sequencers, which are really extremely technical.)

The Virus TI, which promised "total integration" with the sequencer with a VST gimmick, was released almost 11 months late, and for 2 months the mailing lists were FULL, FULL of issues highly similar to the XTC ones. Worked for some people, not for others, works but clicks n pops, works but plugin-delay compensation/latency is unusable, crashes but managed to make it work after a full XP re-install (mm deja-vu =P), OMG it took Access 10 months to get this out the door why doesn't it work well for everyone OMG I'm selling mine and never touching Access products again, etc etc.

So if you want to have some XTC-mode that works, take your issue to Steinberg (who designed the VST protocol), take your issue to sequencer vendors and developers to try and get them to agree on a standard to integrate hardware (they're in competition technically, so good luck!!!), or even to Digidesign, if you bug them they might open up the whole TDM standard, just for you =P But before everyone agrees on a standard, or before Creamware releases their own sequencing package, you can totally forget about it, and there's very little/none Creamware or anyone else can do about it.

I might be wrong on this, but I don't think XTC, dual cpu support, nor more plugins would have helped Creamware at all. What *would* have helped them is to make all plugins crackable (appeals to the huge but very low fund cheapass VST weenies), OR have some brand sticker involving Waves or whoever else all over the place AND include a ProTools-comparable sequencer to solve the whole integration thinger. And charge more!

Finally, I don't know where you got the idea that most hardware made pre-2000 still works fine on modern computers. Most of that hardware *doesn't even have drivers for Windows XP*. But yeah, if you run Windows ME on that new dualcore, it'll work fine

PS Don't listen to dealers, next time contact the company itself =P

On 2005-12-11 00:30, jea wrote:
Hi Symbiote,

dual cpu's have been there since NT4 for consumers. And this is before 2000. Hardware made before 2000 often work well in modern computers too, with or without dual cpu.

I assure you, my dealer told me that Pulsars were "futureproof", and was leading the most leading technology at that time.

And there were some other DSP choices at that time.

I choose Pulsar One.

But the time went, and I bought several other Creamware cards.

I now have a total of 7 cards, including a Creamware TripleDAT ISA.

But computers have been a lot more powerful also, and with the power we have today, sequencers get more powerfull too. But if we can't use our cards together with integration, we are kind of stuck here in single cpu land. I think that if Creamware (to live up to it's name) have to do this now. They should have done it several years ago, to protect their technology cashwise. Imagine what happened if XTC were working, dual cpu, and lots of new plugins were released in 2000-2003.... UAD and TC would have a hard competition, don't you think? And it must be easy to install, like UAD.

Since CW have not dealt with this issue, I am not sure they ever will, I don't think they are capable of it anymore, I don't think they have any qualified programmers to do it.

Otherwise they would have done it in 2000. To protect their business.

Shayne White · Post by **Shayne White** » Sun Dec 11, 2005 3:41 pm

On 2005-12-11 02:55, astroman wrote:

1 - Shayne has 3 cards, with 1 Pulsar one among them. I bet it's that card that 'disturbs' the setup. According to AndreD a Pulsar one doesn't even run on the latest Intel boards - this is about an AMD setup, but extrapolating may apply.

I can remove the Pulsar 1 card and it still crashes when it's only using a Scope card from last year and a PowerSampler board from 2002.

I can remove the PowerSampler board and have just the Scope and Pulsar 1 cards, and it crashes too. I can remove the Scope card and just run the Pulsar 1 and PowerSampler, and it crashes. Haven't tried it with just the Scope board alone, but I have no reason to believe that would work.

I don't know why everyone is arguing about multiprocessing. IT ISN'T MULTIPROCESSING THAT'S THE PROBLEM! I can put the /ONECPU enty into boot.ini and still make it crash. It's some kind of IRQ problem when it's using 24 IRQs, I'm certain of that. That's the only difference between Windows' ACPI multiprocessing driver (which crashes Scope) and Windows' ACPI singleprocessing driver (which works). What else could it be?

Shayne

jea · Post by **jea** » Mon Dec 12, 2005 8:41 am

Well,

I'm sure CW know.