Forums > Jitter

weird hardware (maybe stupid) question

July 24, 2008 | 10:48 am

New computer but unhappy about Max/MSP/Jiter performances.

After i have noticed the bad performance of my patches i have done tests with a lot of help patches on my two computers:

My older laptop :

PowerBook PPC G4 – 1.5 GHz – L2 Cache 512 KB – 1.5 GB RAM – Bus Speed 167 MHz — ATI Mobility Radeon 9700 – 64 MB VRAM
Mac OS 10.4.9
Max/MSP 4.6.3 Jitter 1.6.3
Always worked with no problem with limited possibilities.

My newer laptop :

MacBook Pro Intel Core 2 Duo – 2.5 GHz – L2 Cache 6 MB – 2 GB RAM – Bus Speed 800 MHz — NVIDIA GeForce 8600M GT – 512 MB VRAM
Mac OS 10.5.4
Max/MSP 4.6.3 Jitter 1.6.3
I was expecting a big performance boost and more possibilities. :(

Very weird that there is no performance difference for example with the [jit.gl.imageunit] help patch on the two complete different laptops with the same Max/MSP/Jitter version?!
Also with a lot of other help patches the performance stay the same.

Is these normal? Why do people then buy new computers? :)
Is the NVIDIA GeForce 8600M GT a bad graphics card?
I don’t think so but do i miss something in the set-up?

Thanks for any help!


July 24, 2008 | 7:37 pm

Hello,

Not sure that I could help you but could you tell how do you evaluate the performance (activity monitor? patch?).


July 24, 2008 | 7:55 pm

my intel mpb stomps on most powerbooks i’ve used.
you mention not seeing any performance improvements in the help patches, well help patches are not intended to use as performance tests, so that’s probably your main issue.

issue #2 might be psychological. ostensibly you just dropped a grip on a laptop, and you want to see results for your investment. perhaps you are expecting too much.

issue #3 is not everything is going to see a dramatic speed increase, especially if you’re not doing a lot of opengl stuff. max/jitter is still single-threaded.

as kyred says, how are you evaluating?


July 25, 2008 | 10:14 am

There were some talk about lousy drivers for NVIDEA on OSX, but maybe this has been resolved in the either 10.5.3 or 10.5.4. I haven’t been following the discussions or been using jitter very much the last months, so I can’t tell.

I did some tests comparing a Macbook Pro with an ATI card, and my current Macbook Pro with a NVIDEA card, earlier this year, and concluded that the ATI-computer performed better than the NVIDEA-card. I used the tutorial patch "42jSlab-comparisonGPU.pat" to do the comparisons. At that time I were a bit unhappy with the results, but though that I always could install Windows via bootcamp, if framerates were going to be crucial for me.

I’m sorry, but I don’t have the numbers for the ATI anymore, but with my current Macbook Pro 2.4, with a GeForce 8600M GT 256MB VRAM, I get about 60 fps. That’s in both MAX5 and MAX 4.6, OSX 10.5.4.


July 28, 2008 | 8:28 am

Thanks to all for reply!

Yes, it is true that i have to do better performance tests.

I whas a little bit confused when i tried to get better and more interesting results with an old patch on a better computer but it isn’t possible.
Maybe also the framerate isn’t telling everything!

Is there any good computer performance test for Jitter?

I like the idea that it was maybe "psychological", maybe i can do tests about that. :) But i think it is normal that when you come from a 6 year older computer to a new that you expect a boost in everything that you do.

At least i wanna have the possibility to finish work (patches) that where not possible to do on my older computer.

Thanks, grzz, Bart


Viewing 5 posts - 1 through 5 (of 5 total)