Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[henchman]

Hows it running. How many plug-ins have you been able to run?

[Bill Park]

I don't know what GMs experiences might be, but after a couple of trips back to the manufacturer and several months of wasted time, I threw my dual processor AMD machine away.  Out the door, gone.

I bought a Chris Smith (realcomputerguy.com) Intel with an Intel motherboard, problems solved.

I spent the last couple of days with Sebastian from Sequoia.  I asked him about recommended systems for Sequoia/Samplitude.  His first answer (expectedly) was that we should buy the fastest most powerful Windows-based computer that we could afford.  When pressed, he said that these things change from day to day, but as of that day (yesterday... ) he liked the Intel processor on the Intel motherboard, because of the quality of the Intel motherboards.  He also mentioned that they were quite happy with XP over the other flavors of Windows.

Your mileage may vary.

 Bill

[Ethan Winer]

Bill,

> Intel with an Intel motherboard, problems solved. <

Agreed.

I've always bought computers with Intel CPUs and chipsets, and I have never ever had a problem. Some computer magazines claim that AMD and VIA etc have finally got their act together, but I remain skeptical. Moreover, an editorial I just read (forget where - SOS?) addressed this very issue and said all the pro audio developers test with Intel, and only sometimes try an AMD. So to me the small difference in price between AMD and Intel is not worth the risk of buying a new computer and finding out later - possibly a year later when you get some new program or plug-in - that something doesn't work because of the CPU or chipset.

--Ethan

[bblackwood]

Running XP on an AMD/ASUS combo here and have had zero problems in over a year of every day service...

[oudplayer]

Over here, we've got Samplitude running on AMD Barton 2500+ and Athlon 64 3400+ systems (ASUS mobos) - very fast, and more processing power than we need. Also, have never had a crash of any sort. I haven't tried a test of maximum simultaneous record tracks nor of maximum plugins, but we've had sessions with a number of room-simulator reverbs, eqs and compressors on all 32 tracks (24 bit 96k), and a whole host of Mix Buss and Aux effects without pushing the limits of the CPU load.

Protools 6.1 LE also works great on these systems, as does Ableton Live.

I've heard that the same systems, though, when used extensively for email and web surfing and graphics stuff in addition to audio, start to develop unique "personalities" that aren't desirable in a professional audio recording environment... so make it lean and you'll be much happier. The same probably goes for Intel-based computers, though, too (and probably has a lot more to do with hard drive fragmentation and XP resources than with the chipsets)

[Bill Park]

Please understand... the problems were from jump, and continuous.  I also know many people with rocking AMD systems, and they are quite happy with them.  I went through 2 motherboards and countless other changes, with no joy.

My point, which was similar to Ethan's, is that the developers mostly use Intels (from what they tell me...), and if pressed, they recommend Intel.  You may have noticed that my synopsis of Sebastians' remarks included the motherboard.  This was a point that he was careful to emphasise...  not all motherboards are created equal.  

In the old days, SF used to develop on the Alpha.  This anoyed me because mere mortals could not afford them.  I was a part fo their beta test team.  My P-90 at the time could not begin to perform to the Alpha level, yet it was a top-rated performer in the PC world.  I'm glad that the developers are using what many of us use today.

According to Sebastian, the differences in actual real-world performance between AMD and Intel processors -once installed in a system- are minisule and will likely never be realized or noticed by any end user of audio products.  But the better constructed Intel motherboards are the driver for his recomendation.

He did not say that there were not any good motherboards for the AMD chipset.  What he said was that the Intel mobo with the Intel processor is his recommended system because of the quality of the motherboard construction.  He did not recommend any non-Intel motherboards for the Intel, either.

 Bill

[George Massenburg]

You know, we're doing Nuendo development on this machine, and I haven't really stressed it in the studio just yet.

But I can tell you so far that this is a superb machine - extraordinarily reliable and robust.  We have been running things like a 3D Graphics package on the machine and it runs rings around anything else we've seen.

I'll stay in touch with this.

George

[audio2u]

My 2 cents worth.

My home studio setup is...
AMD AthlonXP 2800+ processor (clock speed is 2083MHz, from memory)
ASUS A7V8X-X motherboard
1GB DDR333 RAM
Windows 2000 Pro + SP4

The setup I use at one of my clients facilities (I freelance too) is...
Intel P4 2.4GHz (not sure if that means clock speed of 2400 MHz or not)
MB unknown
1GB RAM (type unknown)
Windows 2000 + SP3

...and what I can tell you is this...

Running Cool Edit Pro v2.1 with Direct X plugins, the Intel box eats my home systems for breakfast in terms of straight out speed.
But having used both systems, I have no major gripes with either.

[Gognak!!!]

I've been recording on computers since DigiDesign came out with sound tools for the Atari 1040st.  Just so you all know I have used every type of computer available. I have found that sound flounders if it's not a Mac.  There is a huge difference in the way these machines encode audio. Don't be fooled go Mac.  It's not about crashing it's about end results.  Sound. Errors in your audio files
that play back compromised and quantized. Because 256 character machines make more mistakes.  Good luck at any rate...
      George Panagopoulos

[brandondrury]

The AMD CPU / ASUS motherboard rig has worked flawlessly for me for over a year now as well, for whatever it's worth.

Brandon

[atticus]

Gognak!!! wrote on Thu, 29 April 2004 02:53

I've been recording on computers since DigiDesign came out with sound tools for the Atari 1040st.  Just so you all know I have used every type of computer available. I have found that sound flounders if it's not a Mac.  There is a huge difference in the way these machines encode audio. Don't be fooled go Mac.  It's not about crashing it's about end results.  Sound. Errors in your audio files that play back compromised and quantized. Because 256 character machines make more mistakes.  Good luck at any rate...       George Panagopoulos

WHAT?!?! So you're saying that Macs sound better then PC's do? If this is true I'd be really interested in the proof. Thanks.

[graham]

atticus wrote on Thu, 29 April 2004 09:51

Gognak!!! wrote on Thu, 29 April 2004 02:53 I've been recording on computers since DigiDesign came out with sound tools for the Atari 1040st.  Just so you all know I have used every type of computer available. I have found that sound flounders if it's not a Mac.  There is a huge difference in the way these machines encode audio. Don't be fooled go Mac.  It's not about crashing it's about end results.  Sound. Errors in your audio files that play back compromised and quantized. Because 256 character machines make more mistakes.  Good luck at any rate...       George Panagopoulos WHAT?!?! So you're saying that Macs sound better then PC's do? If this is true I'd be really interested in the proof. Thanks.

This sounds rather ridiculous.  Though I suppose you could run some tests with Nuendo or CubaseSX since they are cross-platform.

Graham

[bblackwood]

Surely he's joking fellas.

PCs handle far more important data than audio every day without error...

[graham]

LOL, I guess I should pay more attention to smileys...  and relax.

Graham

[George Massenburg]

Guys,

This is the perfect opportunity for me to say this.

My dear friends, please do not even THINK of starting a MAC vs PC flame war here.  A Mac sounds no different than a PC.  

Period.

Also, do not BEGIN to think of starting a floating point vs fixed dialog here unless you're prepared to do REAL math and some REAL science.

Capiche?

George

[natpub]

George Massenburg wrote on Thu, 29 April 2004 22:26

My dear friends, please do not even THINK of starting a MAC vs PC flame war here.

Thank God  

-KT

[Rob Darling]

Hey guys, in case you didn't notice, this is computer technology we're talking about here.  Unless you are really keeping up on things and testing things, passing on knowledge about experiences from years back just doesn't work.  

Check this thread out if you want to know about the current state of the art.

http://forum.nuendo.com/forum/Forum3/HTML/005409.html

It doesn't get good until Brian T. (Brian Tankersley, well known for always trying to push things to as far as they'll go, but doing it thoroughly) chimes in, but you will learn a lot by reading all the way through.

r.

[tptman]

One of the reasons developers tend to develop on Intel machines is that Intel optimizations are available on Intel CPU's before they are available on any others (of course), and developers are naturally going to want to write code to take advantage of these technologies for the speed gains available.

That being said, I am fairly intimate with this side of the business, as some of you may know, and routinely use several different Windows boxes. I am speaking of current technology, not historical technology. In my experience, there is little or no difference in the reliability of current AMD-Asus combos compared to Intel-Intel.

If we put aside the reliability debate, we are left with essentially two factors that may play into a purchase decision, and those are performance and cost.

Once again, Intel optimizations are a big deal here. New code is going to be written to take advantage of the latest optimizations available. In many cases, developers pass their DSP intensive code to Intel engineers, who then make suggestions on how to better utilize Intel optimizations. IF you are using newer software, and if that software is written to use the very latest Intel optimizations, you will realize a performance gain over the AMD option.

That being said, there are many factors which may cause your favorite software to NOT take advantage of the very latest intel optimizations, in which case, cycle per cycle, the AMD product beats the Intel one on pretty much every level (except perhaps in the case where hyperthreading is advantageous to the process at hand).

As for cost, we are all very familiar with the advantages of AMD.

Really, the reliability problems of AMD machines has little to do with AMD and much to do with OUTDATED motherboards. For someone considering what to buy today, reliability should not enter into the equation. Get a good Asus MB (there are others that are good) with an AMD64 processor, or get and Intel/Intel combo, and you'll see no difference in reliability. You will see a difference in performance/cost, and that's a difference I suspect you'll like.

[natpub]

Yah, I was just going to ask if anyone was familiar with how 64 bit might affect us in the future, as the O/S is updated and software begins to be written for it?

Thx,

[Rob Darling]

Until Microsoft is at least one or two revisions into a 64 bit os, I can't imagine music software really being ready for it (witness the lack of music support in OS X until well into 10.2) How long will that be?  Don't know- two, three years?

In the meantime, the real advantage is that the AMD CPU's are being designed in a way that the PCI bottlenecks of the past can be alleviated deeply.  Also, some of the more troublesome tasks that the motherboard and chipset used to do are now on the CPU, so the mobo will be less of a variable.

[Phillip Graham]

Hello All,

I interned in the AMD Austin wafer fab for most of 2001.  That was known as AMD's FAB25, but now makes flash RAM for Spansion.  I had several friends doing design side work.

AMD's internal reference test motherboards were, at the time, made by Tyan.  The more budget conscious internal people used ASUS at home.

In raw FPU performance the Athlon (i.e. Thunderbird and later) core outpaced Intel's designs.  The last Intel P3 core (Norwood?) had better FPU performance than the P4's of the time.  AMD did consitently lag in implementing all the extra SSE extensions.

What really hurt the Athlon core, other than lower MHz, was lack of integrated thermal protection.  This was more due to great patents on Intel's part than on an inability to implement such things.

What is now the Athlon64 was not yet "taped out" when I left, but I know that they were going to address the thermal issues.

I don't think anyone has ever gotten so much life out of a fundamental CPU design as AMD did out of the Tbird.  Tweaked, prodded, expanded, hot rodded for many years to keep competitive.

Many of the DEC Alpha people are now at AMD, and I think this explains their good chip architectures.  At the time we definitely lagged Intel in terms of our ability to fab some fundamental steps that allow for the high GHz clock speeds.

Being tied to VIA chipsets is definitely an issue for AMD, but they continue to do better and better.  It is a bit nerve wracking when they release a new design.  AMD's simply not large enough a company to be able to do chips and chipsets at the same time.

Anyways, AMD was a wonderful place to intern, Austin's a great town, and I still use AMD processors to this day, even though the Ali chipset in my laptop has crappy USB implementation.

Some perspective