• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Core2Duo mATX vs. ATX MB performance comparisons

For the longest time, I've been looking for some comparison between the Core 2 Duo micro-ATX motherboards vs. their ATX counterparts, for non-gaming tasks such as video / audio editing, general productivity, multi-tasking etc.

Specifically, I'm looking for comparison between the Asus P5B-VM (mATX) vs. Asus P5B-E (ATX).

For example, how much of a performance hit does a G965 type mATX motherboard with integrated graphics incur as a result of sharing memory bus bandwidth with the CPU, for NON-GAMING benchmarks, compared to regular C2D ATX boards?

Parsing for various online reviews across the web, I see a surprisingly big performance gap between the the same benchmarks across mATX vs. ATX motherboards based on G965 vs. P965, 975x, nF680 respectively. Then again, these two sets of reviews almost always seem to have differing benchmark configurations, if even the same benchmarks.

It's surprising and frustrating that Anandtech has YET to post a review focused on mATX based C2D motherboards, when they've been out for months now. How are folks to make an educated decision without proper benchmark data?!


As a last resort in the interim, I'd like to find out if anyone has first-hand or at least second-hand experience / pointers to online reviews for this comparison?

 
Shouldn't be much difference at all. If they have the same chipset, then performance should be the same. Heck, even if they don't have the same chipset, the difference is almost impossible to notice with the eyes.
 
Comparing motherboards is like comparing potatoes at the supermarket. Once you peel them and mash them, they're all taste the same - you'll never know the difference. If you took the fastest chipset motherboard and compared it to the slowest, you would NEVER be able to tell the difference.

The only thing to worry about with motherboards is reliability. Stick with Asus, AOpen, etc. and you'll be okay.
 
aurareturn, actually, that's where things are different. P965 is NOT identical to G965. Most importantly, the latter shares its memory bandwidth between the requests from the CPU 'and' those from the embedded GPU. This is really what I'm curious about - how much of a performance hit does this G965 have for non-gaming tasks.

Fun Guy, I agree that for the most part, for a given CPU family, all the chipsets should provide the same performance. However, I asked this question because I vaguely recall seeing a couple of reviews over the years here at AT as well as other sites that showed noticeable (>5-10%) improvements in certain benchmarks of interest to me (video / audio encoding, USB / Firewire performance etc.), going from one chipset to another for the same CPU family. Can't recall the specifics now.

 
Audio and Video editing shouldn't be done on intergrated platforms that have to share bandwidth. Because you want continuity of streaming memory performance. You CAN actually work on it, but it's not advisable to do so.

At work, all the systems are SFF (Maxt and mBTX) and when they need to do a specific application, we add a PCI-express graphics card to them such as an nVidia Quadro 220. It isn't exactly necessary, but the Dual channel performance need not to be interrupted by intergrated GPU request during application activity.

The P965 and G965 would be the same performance wise if you decide to add a video card to the G965.

For desktop publishing, there is no difference really.
 
Thanks Remedy, that's excactly what I'm looking for more data on. How much performance drop have you noticed on your work systems when using integrated graphics vs. dedicated ones during audio/video editing tasks - say encoding, premiere/photoshop elements etc. (whatever you've tried)?

Do you have any first-hand benchmarks in these categories using your work systems? That would be VERY helpful data to have.

I would drop an all-in-wonder in a blink of an eye right now, if they had one based on the new Theater650 chip. Unfortunately, AMD / ATI has been REEALLY slow to act on getting such an AIW card out the door. Until then, I'd rather not spend 50-100 bucks on a PCIe video card when I don't play games and don't want additional noise / heat in the SFF case. This question is to understand what I'm losing going with G965 until that product comes to market.

Another question: assuming you add a dedicated graphics card, do you have to disable the graphics function of G965 in BIOS explicitly, to truly regain the bandwidth? Or does it stop using memory bandwidth as soon as you stop hooking up a VGA cable to the on-board connector?
 
Back
Top