Gigabyte GA-X38-DQ6: An early look at X38
by Gary Key on September 4, 2007 3:00 AM EST- Posted in
- Motherboards
Memory Testing
The Sandra unbuffered scores on the Gigabyte board are around 4% better at stock speeds and 7% better in our overclocked scenario. In our SuperPI 1.5 benchmark that relies on low latencies for improved scores, we see where this advantage is apparent. We consider these results to be excellent as the X38 chipset, BIOS, and board are engineering samples and we fully expect to see further improvements in the retail boards.
Q6600 9x266 Memory Tests
We were able to set our timings to 4-4-4-12 at DDR2-1066 by increasing the memory voltage to 2.250V on the Gigabyte board and 2.275V on the abit board. In this particular test, our two boards are basically equal with write, latency, and copy tests favoring the X38 and read throughput going to the P35.
Q6600 9x400 Memory Tests
Like our stock scores, the two boards are basically equal with write, latency, and copy tests favoring the X38 and read throughput going to the P35. We set our memory timings to 5-5-5-18, bumped the memory voltages up to 2.275V and 2.350V, and equalized the memory sub-timings where applicable.
 |
| Click to enlarge |
The Sandra unbuffered scores on the Gigabyte board are around 4% better at stock speeds and 7% better in our overclocked scenario. In our SuperPI 1.5 benchmark that relies on low latencies for improved scores, we see where this advantage is apparent. We consider these results to be excellent as the X38 chipset, BIOS, and board are engineering samples and we fully expect to see further improvements in the retail boards.
Q6600 9x266 Memory Tests
 |
| Click to enlarge |
 |
| Click to enlarge |
We were able to set our timings to 4-4-4-12 at DDR2-1066 by increasing the memory voltage to 2.250V on the Gigabyte board and 2.275V on the abit board. In this particular test, our two boards are basically equal with write, latency, and copy tests favoring the X38 and read throughput going to the P35.
Q6600 9x400 Memory Tests
 |
| Click to enlarge |
 |
| Click to enlarge |
Like our stock scores, the two boards are basically equal with write, latency, and copy tests favoring the X38 and read throughput going to the P35. We set our memory timings to 5-5-5-18, bumped the memory voltages up to 2.275V and 2.350V, and equalized the memory sub-timings where applicable.
Facebook
Twitter
RSS
26 Comments
View All Comments
jay401 - Tuesday, September 4, 2007 - link
How does it compare to the previous gen motherboards almost everyone is running?Is it really worth an upgrade from 650i/680i/P965/975X?
That's really what matters.
TA152H - Tuesday, September 4, 2007 - link
The chart "Media encoding - Sony Vegas" makes no sense. I think you got the stock performances twisted.Also three to five percent over the P35 sounds incredible, as in not credible. Especially with the Penryn (with it's bigger cache). When the motherboard makers talk this, you should try to find out what processor they are talking about. Maybe on a 512K processor it can reach it, but on a 4 MB processor, on most benchmarks, it doesn't sound realistic at all. Well, it also depends what they are comparing it to. It says the P35, but maybe they meant the 975X, which is much more likely. Crossfire, of course, will be much higher, but how many people are actually using this? It's a very, very low percentage. So, I think people are going to be disappointed again, like they were with the Penryn, because of misguided expectations. Penryn, like x38, is a great product, it's a shame people lose that because of these unrealistic expectations. Still, anything is possible, but if it averages 4% on most applications against the P35, on a Penryn (will enthusiasists buy the Conroe when it comes out?), I'd be very surprised.
One suggestion, when you do the final review, run it with a Conroe-L as well. Why would an enthusiast site run this as well? Well, if I needed a machine, right now (as in I had one computer and it died), I would buy it for $40 and then buy a Penryn when it became available, or cheap, and then use it in a low power computer. You might see the 3-5% claim have a chance on a small cache processor. That's probably what they were talking about, without actually wanting to say it.
phusg - Tuesday, September 4, 2007 - link
Sounds an interesting upgrade path. But aren't Conroe-L's the Celeron 4xx's which are socket 478? Do we even know what socket Penryn will be?
TA152H - Tuesday, September 4, 2007 - link
Conroe-L is LGA 775, same as Penryn will be. I suspect almost any motherboard being made now will work with the Penryn, or whatever the desktop version is called, really. I can't keep up with these stupid names AMD and Intel come up with for their processors. The ass that thought of Barcelona should be sterilized. What an absurd name. If they had any brains they might make code names and add something like -L for light, -M for mobile, -S for server, -QM for Quad Mobile, etc...Instead, they have these horrendous code names, which are suitably replaced by horrendous model numbers. I really can't keep track. Pentium III 1 GHz was so much easier to remember.
Guuts - Tuesday, September 4, 2007 - link
Gary,Page 2: Third paragraph from the bottom, should have (DDR2-1066) and not (DDR-1066).
Page 3: Missing graphic at the top of the page, text starting off with "memory-lg.png memory-sm.png" is probably the cause?
Page 5: Chart 2 (Sony Vegas) either has the stock speed bars' labels reversed, or your text summary is incorrect, as the chart doesn't show the Gigabyte board ahead in both stock and overclocked speeds.
Page 7: Second paragraph, last sentence, "...better than several retails boards" should just be "retail".
Good article overall, though I was expecting a bit more of an improvement over the P35. I bet there will be a nice boost over the 975X when we see the full review, however. Thanks for the early look, Gary.
Now...where is the P35 board roundup? :-p
JarredWalton - Tuesday, September 4, 2007 - link
Hi - sorry about the graph missing... that was my mistake. You can blame me for a few other missed typos as well if you'd like. I'm not sure on the Sony Vegas chart if there's some sort of labeling error or if we need to correct the text, so I'm leaving it alone.--Jarred