PT8 CPU Usage

[Driven01]

You can add me to the list of CPU eating Pro Tools 8. Nails mine at nearly 100%. Definately a GUI Issue. Opened, 95% CPU. Minimized, 3%. C'mon Digi! How about a fix!!! Paid over 200 bucks for something I can't use.

[Ling]

I'm having this problem too.

Since not everyone is having this...is there a specific combination of hardware that's doing this? Is it a combo of certain mobo chipsets with certain video cards? Anybody know?

btw - I swapped video cards, nVidea to ATI - it didn't change anything...tried the card's drivers, Window's default drivers...nada. (acceleration set to none in all cases)






Processor: Intel Q9550, 2.8gig
Mobo: Asus P5K-WS, w/ P35 chipset
Ram: Corsair Dominator 4g (2x2gig)
Power Supply: Corsair 520HX 520watts
Drives: 3x WD caviar SE16 WD7500AAKS 750gig
Video: Asus EN9600GT Nvidia, 2DVI,
(alt: PowerColor ATI radeon 4650, 512m, 2dvi, fanless)
no noticable difference in performance between the 2 vid cards
Cooler: Thermalright Ultra-120 eXtreme
Firewire card: ADS Pyro PCI 64R2
Interface: Profire 2626 firewire
M-Powered PT8-CS2

[RoMi]

My configuration:
Intel quad Q6600
ASUS P5K DeLuxe WiFi (disabled)
G.Skill 4GB (2 x 2GB DDR2 1066)
Radeon HD 2600PRO 512MB 128-bit
Bunch of internal (SATA) and external (eSATA, USB & Firewire) HDD’s
OS+programs and media files on separate HDD’s (of course)
ARK 4U-500-CA Black 4U Rackmount Case
MoBo onboard Firewire or Digi aproved PCI Firewire card (same results)
002R
PT8 (on one partition)
PT7.4 (on the other partition)
...
Different configs - same problems...

I like the "Driven01's" description "...CPU eating Pro Tools 8..." - it sounds so SiFi

[Driven01]

Yup, I tried everything possible to correct the issue within my laptop to no avail. So I figure PT 8 just likes to chew away on my cpu. I really wish it wouldn't. Munch, munch, click, pop, cpu usage is too high, you'll have to stop. As funny as it sounds, it's pissin' me off. Especially after finding out that this has been going on pretty much since 8's release. I bought PT recently as an "upgrade" from Cubase LE. So far it's been a downgrade. Pretty sad considering Cubase LE is a "lite" version of Cubase.

So come on Digi. How 'bout a fix? Huh? Please? Pretty please? I know this very shapely blonde who'll...

[DigiTechSupt]

Originally Posted by Driven01

Yup, I tried everything possible to correct the issue within my laptop to no avail. So I figure PT 8 just likes to chew away on my cpu. I really wish it wouldn't. Munch, munch, click, pop, cpu usage is too high, you'll have to stop. As funny as it sounds, it's pissin' me off. Especially after finding out that this has been going on pretty much since 8's release. I bought PT recently as an "upgrade" from Cubase LE. So far it's been a downgrade. Pretty sad considering Cubase LE is a "lite" version of Cubase.

So come on Digi. How 'bout a fix? Huh? Please? Pretty please? I know this very shapely blonde who'll...

Everyone - I know this is a very frustrating issue and we're working on it as quickly as we can.

If I can take a moment to explain a little about the production cycles and why things 'take so long', maybe it will shed a little light on things.

When a problem is identified, a bug is logged and assigned to an engineer to be looked at. They are typically done in order of importance, but also factors like complexity, invasiveness and the current queue for that developer have an effect on how quickly they can do anything about it. They may be in the middle of another (or multiple) other fixes and sometimes it's not practical to drop everything in favor of working on a specific bug - mainly because code can be extremely complex and working out a difficult problem could mean managing multiple logic trees that, if the engineer turns to another problem, will need to be re-traced again (at a loss of time) once the engineer comes back to it. Engineers typically also have a 'specialization' in a particular code area and sometimes it requires the 'best guy' to work on a particularly difficult fix.

Once the engineer is on the case, they have to be able to reproduce the problem - sometimes this can be intermittent and they have to take the time to 'trap' or write code that isolates the issue and identifies what's going on at that moment. After identifying the problem, the code needs to be gone through to determine what's happening that's causing the problem, then re-coded to fix it. Sometimes it's easy - one line. Most of the time, though, it's much more than that and sometimes 'bleeds' over into other code areas, all of which need to be traced to find areas that code changes could be affecting and make sure that a fix for one area isn't going to break something elsewhere. When problems become complex, trapping, identifying, re-coding, testing and re-coding where needed can take several weeks or even months - this particular issues is one that's rather complex and has been difficult in almost every area. It's intermittent, difficult to trap, code changes affect multiple areas and testing has been shown to improve significantly on some systems, not as much on others. Because of the variety of hardware available on PC's and how that hardware affects the code, there has been several areas that have to be tweaked to accommodate those differences and that's taken even more time to identify and re-code. Then everything has to be retested again. Lather, rinse, repeat.

I don't know if knowing any of this really helps anyone - I'm just trying to give a realistic idea of what a typical 'cycle' is for bug fixes.

[Greg555]

I got in a hurry when I posted my problem like this. And didnt search for related ones then I found this thread. Anyway my rescources meters in PT 8 show nothing, and the xp performance show CPU at 100% consistantly it never goes under 100% my system idle rescources is showing zero. but odly my session still plays but sometimes the program locks up.

Greg

[idylldon]

Originally Posted by DigiTechSupt

I don't know if knowing any of this really helps anyone - I'm just trying to give a realistic idea of what a typical 'cycle' is for bug fixes.

I appreciate your thorough explanation, and I look forward to trying out the fix once it's available.

Cheers,
--
Don

[Ling]

Thanks Digi - it's good to have a word from you.

[Driven01]

Thanks for popping in and letting us know that the issue hasn't been buried and that the engineers are truly working on it. Your explaination of the "cycle" is helpful and highlights the apparent complexity of this particular issue. If it were a "one liner" it'd been fixed already.

What would be of more help is that there's some sort of semi-realistic projection of when Digi feels the fix might be available. A lot of people are either "hanging" on to their current PT8 installation hoping on an imminent fix release, uninstalling 8 and reverting to 7.4 because they simply can't wait, or getting totally cheesed off and dumping PT altogether for another product that performs the way they expect it to.

It'd also be nice if you guys could pop in every couple of weeks to give us all an update of where you're at in the process. This would let all of us know that 1.) Yes, they're still working on it, 2.) How far along in the process the issue is., and 3.) When we're getting close to a fix. I personally think that this alone would give us all a bit of hope and peace of mind.

I might also suggest that you make this thread a sticky so that 1.) Users can quickly find the issue and know that they are not alone, 2.) post their Sandra reports, etc. to help the engineers pinpoint the problem area (although based on the miriad of PC configurations it has affected there doesn't appear to be much in common for anyone particular problem area related to PC's), and 3.) Centralize the problem in the DUC There's a boatload of related threads. I know it wouldn't be a pretty thing to have at the top of the DUC but it may help relieve some user related stress.

Thanks for keeping at it and we look forward to a fix. The offer of the Blonde still stands.

[Bob Olhsson]

I'm sure if it were easy, it would have been fixed weeks ago.

Better to have it fixed right than deal with a quick fix that only breaks something else.