FIRST LOOK: ULi M1695 PCIe/AGP Socket 939 for Athlon 64
by Wesley Fink on July 13, 2005 12:05 AM EST- Posted in
- Motherboards
tRAS and Memory Stress Testing
Memory tRAS Recommendations
In past reviews, memory bandwidth tests established that a tRAS setting of 11 or 12 was generally best for nForce2, a tRAS of 10 was optimal for the nForce3 chipset, and a tRAS of 7 was the best choice for the nForce4 chipset. Since this is our first review of a ULi chipset, tRAS timings were first tested with memtest86, a free diagnostic program with its own boot OS that will boot from either a floppy disk or optical disk. Bandwidth of OCZ PC3200 Platinum Rev. 2, based on Samsung TCCD chips, was measured from tRas 5 to tRAS 13 to determine the best tRAS setting for the ULi chipset.| Memtest86 Bandwidth ULi M1695 with Athlon 64 4000+ |
|
| 5 tRAS | 1874 |
| 6 tRAS | 1913 |
| 7 tRAS | 1953 |
| 8 tRAS | 1953 |
| 9 tRAS | 1994 |
| 10 tRAS | 1994 |
| 11 tRAS | 1994 |
| 12 tRAS | 1874 |
| 13 tRAS | 1974 |
The best bandwidth was achieved with this combination of ULi M1695/4000+/TCCD in the 9 to 11 range, so a mid-value tRAS of 10 was chosen for all tests. It appears that optimal tRAS timings may also be memory dependent, so we recommend a quick series of memtest86 to establish the optimum tRAS timings for other memories on the ULi chipset.
Memory Stress Test
Our memory stress test measures the ability of the ULi to operate at its officially supported memory frequency (400MHz DDR), at the lowest memory timings that OCZ PC3200 Platinum Rev. 2 modules will support. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory. To make sure that memory performed properly in Dual-Channel mode, memory was only tested using either one dual-channel (2 DIMMs) or 2 dual-channels (4 DIMMs).| Stable DDR400 Timings - One Dual-Channel (2/4 DIMMs populated) |
|
| Clock Speed: | 200MHz |
| CAS Latency: | 2.0 |
| RAS to CAS Delay: | 2T |
| RAS Precharge: | 10T |
| Precharge Delay: | 2T |
| Command Rate: | 1T |
Using two DIMMs in Dual-Channel 128-bit mode, the memory performed in all benchmarks at the fastest 2-2-2-10 timings at default voltage, which was the only memory voltage available.
| Stable DDR400 Timings - 4 DIMMs (4/4 DIMMs populated) |
|
| Clock Speed: | 200MHz |
| CAS Latency: | 2.0 |
| RAS to CAS Delay: | 2T |
| RAS Precharge: | 10T |
| Precharge Delay: | 2T |
| Command Rate: | 2T |
Tests with all four DIMM slots populated on the ULi M1695 required a 2T Command Rate with 4 DIMMs in two dual channels. This is the pattern seen on other top-performing Socket 939 boards. However, the ULi had no problem running at a 200 CPU speed setting with 4 double-sided DIMMs. This performance is certainly competitive with the best that we have seen on nForce4 motherboards for Socket 939.
Facebook
Twitter
RSS
72 Comments
View All Comments
jpkomm - Tuesday, August 2, 2005 - link
In reference to the comments on ASrock's making of this board, I got word back from their US sales division as to when their board will actually be released in the US market. They said they have no plans on selling the "939Dual-SATAII" motherboard in the US region. Oi that bites. Anyone know of other companies planning on producing these boards? The ASrock board looks and (from the reviews) performs great, but I'm not going to hold out for the chance that it never comes to the US market.justly - Wednesday, August 3, 2005 - link
Are you actually surprised by this? I would probably do the same thing since most Americans will only purchase a product surrounded by marketing hype.
ElJefe - Monday, August 22, 2005 - link
the dual sata 939 will be sold in the US. asrock said it, asrock america sales said it too................(just in case anyone reads this thread... but the real discussion is in the 2nd, newest article here at anandtech)
hazeldene - Tuesday, July 19, 2005 - link
I wonder if you would be able to use an ATI AGP X-Series card as a Slave with an ATI X-series crossfire in the PCI-express slot... that would be the cheapest Crossfire / SLi upgrade ever !!!mistersnail - Monday, July 18, 2005 - link
http://www.hkepc.com/hwdb/m1695-uli-1.htmCrossfire can be done on this board
val - Sunday, July 17, 2005 - link
#65 last but not least. Suyin is usually the very last thing company does. So buy it until you can. We loves all of you who look to only syntetic tests and benchmarks, we have cheap intel cpus - thanks to you .val - Sunday, July 17, 2005 - link
#65 if you have in your room already 35, than there is something wrong with your room planing. No pcs can change it if you are still running heating or teaching in stove. And saying it last, do your math and learn what percents and thermall loss means. The difference to AMD platform is in percents of complete amount of heaters in room small and thermal loss increases exponentialy.However, world proves that some heating issues cannot change the fact that no "fanatic - glad of self punishment" admin would install room of amd systems.
mino - Saturday, July 16, 2005 - link
#62 it was not mentioned exactly anywhere (no one knows it actually) but rough estimate is 4 to 8 weeks. I believe it will be much closer to the seconf number.#63 One think I forgot something. Do you really believe it is technically feasible to guarantee 38C ambient temperature when the is about 35C in the room ?
mino - Saturday, July 16, 2005 - link
1) Actually I read tom's since they make some unique test sometimes. They are one of(many) relatively goog sites.2) Cooler classroom _will_ help mostly better learning/teaching, I said that.
3) Actually I like AMD's approach. However where bussiness is concerned only facts matter. I bought many Intel systems recently, but they were chosen because of their better suitability then. This is not the case here (meaning almost whole [prescott lineup).
4) actually in the room is normally running _single_ central heater. You forgot that rooms are not allways build in separate building. In this case only contacts with the outside are windows. Also remember the radiation form of heating causes lower temperature than convection at the sam output. The out pu t during the day should be around ~200W/machine(incl.LCD) + 1kW of lights make ~7kW which is close to your number. This however considers idle state, but there are working people about 50-70% of time so real average would be about 8kW for prescott room.
5) Anobody saying idle Prescott PC+LCD will consume about 100W of power is either mislleading or has no idea what he is tlaking about. This figure however fits our Newcastle's nicely.
6) Yes anybody could suck any number of any claims out of its finger anytime. Nothing new here.
What matters to me our academic society could work in these new classrooms without major health issues and having to abuse IT staff for what they bought.
That is what make me happy no matter what you think.
val - Saturday, July 16, 2005 - link
#60: it is whole crap.You are reading too much toms hardware. Intel is not throttling when you install cooler properly.
Cooler classroom will not help to admin when he have to spend there every day one hour.
You are talking like typical AMD fanboy, using untrusted claims which nobody would ever believe. Trust me, that 5100W running 8 hours a day will not heat up to 30° when outside is -10 with "any" isolation.
And even if it would, it would not heat it much more than 3000W using any power saving PC.
I can suck out of my finger 20 claims like that one you posted - if you want, so save your time writing stories.