uATX Results

[JKing76]

There are so many cheap, passively cooled entry level video cards that I don't think I'd use an integrated graphics chipset in a machine I'd ever want to upgrade.

 

[Gary Key]

Intel Update on G965/G33 XP drivers, but no performance updates for Vista yet

Our Editorial Calendar underwent a couple of changes today due to article changes for tomorrow and Wednesday. The first uATX board overview goes up on 8/23 in the afternoon or that night for Friday the 24th, the next part is up on 8/29 followed by another one on 8/31. Somewhere in the middle of all that is the first of two P35 articles and the start of the GPU/IGP image comparisons.

I will release any updated BIOS spins on Wednesday for the boards we tested. The SB600 equipped boards are all undergoing BIOS changes as there were several compatibility issues with the new Blu-ray drives from Pioneer and Lite-On.

 

[FireTech]

Originally posted by: JKing76
There are so many cheap, passively cooled entry level video cards that I don't think I'd use an integrated graphics chipset in a machine I'd ever want to upgrade.

I take it your still talking about the Gigabyte?
OK, so you plug the graphics card into the x16 slot. Where does the x1 tv card have to go?

Originally posted by: Gary KeyOur Editorial Calendar underwent a couple of changes today due to article changes for tomorrow and Wednesday.

Thanks for the update Gary, at least we're being kept informed about the situation now. :thumbsup:

 

[JKing76]

Originally posted by: FireTech

I take it your still talking about the Gigabyte?
OK, so you plug the graphics card into the x16 slot. Where does the x1 tv card have to go?

Huh? The Gigabyte has integrated video. I wouldn't put a video card in it at all. I wouldn't ever upgrade it, either. As a very specific-purpose machine (HTPC), if it's good enough for now it'll be good enough for as long as it takes for ALL components to be outdated and need replacing. If I was building a machine that I might need to upgrade as new tech came out, I wouldn't use an integrated chipset mobo in the first place.

 

[tcsenter]

One of the most common requests we have received is to see how an integrated graphics chipset on the typical µATX platform compares to a more performance-oriented graphics chipset designed for mainstream configurations. We're not trying to turn our µATX roundup into a graphics card shootout, but we decided to look at solutions that range in price from $60 to $125 to see just what type of performance and feature improvements an extra week's paycheck could buy.

Here we go...exactly the kind of spurious comparison I recommended against in response to Gary's solicitation of suggestions for the mATX roundup:

How about testing integrated graphics against sensible entry-level discrete GPUs? e.g. GF6200, 7300GS, RADEON X300, X600.

None of this "Well golly, none of our [free] integrated graphics competitors remotely approach the peformance of the $300 value segment 8800GTS."

Obviously, I used the ~$300 8800GTS as an exaggeration of the pointless comparisons some review sites have made in testing IGP against discrete GPUs they deem to be "mainstream" performance cards that cost as much or substantially more than the motherboard alone. However, the price-performance point at which such comparisons clearly become spurious is actually much less.

$60 ~ $70 is the entry point for cards like GF 7300GT and X1650PRO, with 7600GS available for a mere $15 more, all of which we know will trounce any current IGP by triple digit percentages and it just gets worse from there. $125 is squarely into GF 8600GT territory. This is not unlike comparing how the 'free' stereo speakers built-in to an LCD monitor compare with 5.1 surround systems priced between $60 ~ $125. But hey, why stop there...

- How good are the OEM PSUs bundled with PC cases compared to aftermarket PSUs costing $60 ~ $125?

- How does MS Paint compare to applications such as Corel Paint Shop Pro or Adobe Photoshop Elements?

Cripe! We already know beyond any doubt that most low-end GPUs out there under $60.00 readily out-perform current IGP, some by large margins. Heck, based on the following 3DMark scores, I'm fairly confident my six year-old Radeon 8500 128MB AGP 4x with default clocks (275/275) would make a fair showing against current IGP in both DX8.0 and DX9.0 (using DX8 path) benchmarks:

3DMark01SE = 9000+
3DMark03 = 1300+

This was using S478 P4 2.8GHz (533FSB) on a crappy ECS PM800-M2 motherboard + single channel 512MB DDR333 RAM! I can't imagine the R8500 scoring any worse on a mid-range AMD64 X2 or Intel C2D + dual channel 2GB DDR400 or DDR2-667 RAM.

Including entry-level GPUs in the $30 ~ $60 range such as GF 6200TC, 7200GS, 7300LE, 7300GS, Radeon X1050 and X1550 would have answered any question about the relative performance of everything above those cards. By neglecting this lower-end, it leaves open the question of just where the price-point begins to clearly differentiate integrated and discrete graphics in performance. IOW, how much will it cost to obtain 2x...3x...4x performance gain over IGP?

There are already plenty of reviews of the 7600GT, X1650PRO, and higher GPUs for those who want to know how those discrete GPUs perform relative to other discrete GPUs. Comparing these discrete GPUs to IGP is purely an academic interest of no practical value.

 

[FireTech]

Originally posted by: Gary Key
The first uATX board overview goes up on 8/23 in the afternoon or that night for Friday the 24th,

It's Friday night here and Friday morning Stateside so part 1 can be only minutes away... :laugh:

 

[Skott]

Its Friday 5:06pm EST here. Not surprised it hasnt shown up yet.

 

[FireTech]

Originally posted by: Skott
Its Friday 5:06pm EST here. Not surprised it hasnt shown up yet.

but Gary said it would.... :Q :brokenheart:

 

[Binky]

I call shennanigans! :thumbsdown:

 

[Mr Vain]

I'm waiting for DEANSDAD to comment on this latest twist, it will be a ripper. :evil:

 

[imported_Kreed]

Guys, while we're waiting for the article, what do you think is the best AMD based mATX motherboard?

From what I can gather, the top two contenders are:

Gigabyte GA-MA69GM-S2H (AMD 690G Chipset)

and

Abit AN-M2HD (Nvidia 7050-630A Chipset)

The Asus M2A-VM HDMI is probably a close runner up.


As far as Intel is concerned, the Gigabyte GA-G33M-DS2R seems to be the favourite with the Asus P5K-VM being a close runner up.


Is that a fair assessment of the current state of the Micro-ATX motherboards?

Looking forward to the next generation, we have launches for:

- Nvidia MCP78 IGP chipsets in November 2007.
- ATI RS780 IGP chipsets in Q1 2008.
- Intel Eaglelake-G IGP chipsets in Q2 2008.

All three are claiming full H.264 acceleration and support for DisplayPort.

We also have:

- Intel G35 IGP chipsets due in Q3/Q4 2007
- Nvidia MCP73 IGP chipsets due in September (25th) 2007

However, I consider these as being part of the current generation of IGP chipsets.

 

[Zap]

Originally posted by: Kreed
what do you think is the best AMD based mATX motherboard?

Gigabyte GA-MA69GM-S2H (AMD 690G Chipset)

I'd say maybe this board, based on one review.

 

[Skott]

Biostar has one that seems pretty popular as well. Not sure which AMD mATX mobo is top end right now. Btw, the latest DS2R bios removed the 1:1 divider I'm hearing.

 

[FireTech]

Originally posted by: Mr Vain
I'm waiting for DEANSDAD to comment on this latest twist, it will be a ripper. :evil:

No offence to deansdad (I'm sure it will be a ripper) but I'm actually waiting for Gary to comment on this latest...

 

[firewolfsm]

and we still don't a have a good P35 roundup on this site.

 

[FireTech]

It's a little late but at least it's here! :beer:

 

[JKing76]

Wow, so a dedicated video card is faster than an integrated solution. I'm shocked! In my eyes at least, the uATX review still hasn't started. How long it takes to see some articles about how the board would actually be used?

 

[Zap]

Originally posted by: Gary Key
The first uATX board overview goes up on 8/23

Only a 5 day delay instead of a 5 month delay this time, and no G33 (except for video performance). :thumbsup:

Time to read.

 

[Skott]

Personally, I'm surprised. I really didnt believe it get up this week at all. The info provided so far is nice but way way late. It would have been really nice to have this info months ago concerning the ABIT F-I90 HD. A lot of people needed it then. Its a real shame the Asrock mobo doesnt overclock that well. Maybe the new bios will change that? It would have made a nice alternative to the ABIT.

Hopefully the rest of the round up review wont take days or weeks to get published. It'd be really nice to see it all go up this week and not spread out over the upcoming weeks. (hint hint)

 

[deansdad]

Aldous Huxley said it best when he said: "Facts do not cease to exist because they are ignored."

 

[FireTech]

Originally posted by: Gary Key
The first uATX board overview goes up on 8/23 in the afternoon or that night for Friday the 24th,

It actually made it's appearance sometime on the 28th (Oz time that is).

the next part is up on 8/29 followed by another one on 8/31.

Which even allowing for the compounded delay from part 1 means part 2 is now late, as will be part 3. Got to admire the consistency with Anandtech...

Somewhere in the middle of all that is the first of two P35 articles

It's good to see the P35 guys are now getting fed the same stories we have and are feeling similarly about it.


From the p35 round-up thread>>>>

This thing better be one hellava roundup. My God, this has taken so so long. And tomorrow is the 31. I say there is at least a 50% chance we won't see part 1 for another few weeks.

Whelp, they got me to hit their front page and ad banners every day for two weeks strait. Guess I'm a sucker...thanks for nothing.

 

[imported_Kreed]

I just hope Parts 2, 3 and possibly 4 are posted before these boards become irrelevant. Perhaps Gary will surprise us and post them all in one go, or is that wishful thinking?

 

[Skott]

You know what I'm curious about? Why? We heard some excuses from Gary but it seems there is something going on behind the scenes we dont know about. I mean if the review is done like its been said then why hasnt it been posted? Is it Gary? is it AT higher powers that be? is it some third party like the mobo manufacturers telling AT to wait or else? Is it aliens from another world just enjoying messing with AT users? What/Why is the hold up? Thats what has my curiosity going.

 

[imported_Kreed]

We can pretty much eliminate the possibly that parts 2 and beyond won't appear early next week as it will be dominated by articles on AMD Barcelona. That means if we don't get part 2 today or tomorrow (Friday), we will probably not get it till late next week at the earliest. On top of that, I believe we also have the Intel X38 chipset being launched on September 9th. I am pretty sure AT will give priority to any article on the X38 chipset as (1) it will perceivably have a larger reader base and (2) it's Intel - enuff said. As such, resource is probably being allocated to these "more important" articles, resulting in delays to the mATX mobo reviews.

I understand that this is a business and AT needs to make money. Articles on mainstream products will obviously generate more money than articles on hobbyist products. If the rest of the articles are not going to appear anytime soon, all that I ask is for Gary (or Anand himself) to give us an update (say once a week).

 

[imported_Kreed]

Edit: Double Post