CGTalk - Need advise please

rrsolis

01-04-2005, 07:09 PM

Hi, I'm about to build a system, which will be mainly used to do lightwave modelling, animation and rendering.

I have two options:

a) Intel 3.4 ghz Northwood, 875 intel motherboard, 2 gigs of mem pc3200, Ati 9800 pro (agp)

b) Intel 3.6 ghz Prescott, 915 gigabyte motherboard, 2 gigs of mem pc3200 (not ddr2, as the motherboard supports the old ddr and the new ddr2), msi geforce 6600 (pci express).

Suppose that price is not an issue. What I'm looking for is for advise from the performance side of things. The main reason I'm asking this is because I heard the the northwood processor is faster than the prescott at the same clock speed. But what happens when you add other components, like a faster graphics card? (I Know that right now AMD would be a good choise, but I have a video editing card that does not support amd harware)

Thanks!

brzilian

01-04-2005, 07:18 PM

(I Know that right now AMD would be a good choise, but I have a video editing card that does not support amd harware)

Thanks!
If you are referring to Matrox, I suggest you double check that. Last I heard they now support AMD hardware with their video editing hardware.

MattClary

01-04-2005, 07:31 PM

I would go with an AMD64 CPU and I would go with the best consumer level nVidia card you afford.

rrsolis

01-04-2005, 07:34 PM

You are right, I have a matrox board, but anyway, it is easier for me to get intel compatible hardware than amd's, and the vendor has real good prices on intel stuff

lots

01-04-2005, 09:17 PM

I'd go with option A and a Geforce 6800GT insted of the ATI card :)

rrsolis

01-04-2005, 10:17 PM

Ok, but why would you go for option A? Is it because of better performance compared to the other option or is it because of price?

By the way, this is a reply l got from another forum:

Option B for sure - especially if you are doing real-time 3D rendering. The 6600 will trample all over the 9800 Pro >> The main reason I'm asking this is because I heard the the northwood processor is faster than the prescott at the same clock speed.

Not necessarily - depends on the application and the rest of the components. Pure CPU performance wise, Northwood is faster when the 512 kB cache is not limiting it. If cache size becomes the main bottleneck, the Prescott's double amount of cache will help. But either way, performance difference will be negligible in general.

>> But what happens when you add other components, like a faster graphics card?

If you're gonna do professional 3D rendering, you want the highest end video card you can get your hands on. Saying the applications/code you will run depends on OpenGL/DirectX or any other Graphics API that uses hardware acceleration, then their speed directly depends on the GPU being used (of course, it will also depend on other stuff like how much graphics memory you have, amount of RAM, CPU....it would be pointless to couple the most power GPU with a system that only has 128 MB of RAM for example)

l don't do real time 3d. Any thoughts?

CGTalk Moderation

01-20-2006, 09:00 AM

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.