First AMD Benchmarks With DDR3 Memory Posted

[ShadowFold]

Hold on a sec, why are they using Wprime 1.55? WPrime 2.00 was released Sept 2008.

It's tomshardware funded by Gigashit. What else do you expect?

 

[Octavean]

exactly, thats because there isn't any noticeable gain in performance with any new DDR jump.... and there won't be until they can figure out how to bring down latency.

My point, in case it was missed and my guess is that it was, simply is that the transition from DDR2 to DDR3 is the direction that the industry is going in. The transition from DDR1 to DDR2 was a certainty, why fight it?

For example, some people may want a super fast high end single core processor but Intel and AMD will be giving us Dual Core and Quad Core processors for these markets. Both companies will likely phase out even Dual Core,….

That’s life, just be thankful there isn’t a significant performance deficit in most cases with the new memory type,….. Also be thankful that there is a choice between DDR2 and DDR3 for the Phenom II,….. Lord knows Core i7 users don’t have that choice

 

[EastCoasthandle]

If it's possible, one should just get DDR3-1600 at CL7 and try to get the best speed possible. Again, if the setup will allow you to boot with DDR3-1600 at DDR3-1333 (for example). Get the best stable OC for your PII and the highest possible speed for your ram.

Granted we will still need to see a full review between DDR2 vs DDR3. But if I one was to get a AM3 setup regardless, this would be the route I would take if the ram can be used.

However, another question arises from this. If you are not able to get DDR3-1600 can you do CL5 or CL6 at or around DDR3-1400 (for example)? Time will tell either through reviews or through personal experiences.

 

[imperialreign]

exactly, thats because there isn't any noticeable gain in performance with any new DDR jump.... and there won't be until they can figure out how to bring down latency.

latency timings are not 100% of the performance equation when it comes to the next DDR series.

I'm not going to go into it here, as there are more than enough threads running around TPU that discuss this little aspect; there's no need to get hung up on the high-timings of DDR3 when you stop to consider how much more information DDR3 moves per clock cycle compared to DDR2.

 

[iamverysmart]

exactly, thats because there isn't any noticeable gain in performance with any new DDR jump.... and there won't be until they can figure out how to bring down latency.

The latency is fine. Do some research. The timings don't directly convert to latency.

For example you can have 1066Mhz C6 with lower latency then 800Mhz C4.

 

[EastCoasthandle]

What research? There are no benchmark results using AM3 yet. What we are looking for is to see what results are when AM3 is put to the test. At the time of this post, I don't know of any link providing this information. Do you have a link providing these results using AM3?

 

[kid41212003]

However, another question arises from this. If you are not able to get DDR3-1600 can you do CL5 or CL6 at or around DDR3-1400 (for example)? Time will tell either through reviews or through personal experiences.

Check the current market, there is no such DDR3 at that latency.

But here, the fastest DDR3, and it's not released yet.

It's tomshardware funded by Gigashit. What else do you expect?

GIGAshit? Lolz

At least, GIGA Boards don't need to boot from any kind of devices to update their board bios, (Press End when POST, and update your bios from FAT32 partition), and that isn't from Windows.

 

[EastCoasthandle]

Check the current market, there is no such DDR3 at that latency.

But here, the fastest DDR3, and it's not released yet.

I honestly don't think you understood my question. I am talking about when folk do us DDR3-1600 and are not able to actually get 1600 MHz (something lower around 1400 for example). Will they be able to use CL5 or CL6? This type of question cannot be answered until they are made available for use.

 

[imperialreign]

I honestly don't think you understood my question. I am talking about when folk do us DDR3-1600 and are not able to actually get 1600 MHz (something lower around 1400 for example). Will they be able to use CL5 or CL6? This type of question cannot be answered until they are made available for use.

From what little I've seen, timings with DDR3 don't seem to make too phenomenal of a difference in terms of overall performance. DDR3 seems to benefit a lot more from higher clock speeds than it does tighter timings.

 

[EastCoasthandle]

From what little I've seen, timings with DDR3 don't seem to make too phenomenal of a difference in terms of overall performance. DDR3 seems to benefit a lot more from higher clock speeds than it does tighter timings.

IMO when it comes to ram speed is everything at a decent timing. I was speculating on a specific scenario were a user wasn't able to achieve DDR3-1600.

In the end, we will have to wait and see if they are able to actually get DDR3-1600 stable or not.

 

[Formula350]

On my Toledo 939 X2, the overclock HELP my latency. I get 47ns @ 450mhz mem and 2.7ghz CPU, but 49.9ns @ 426mhz mem and 2.55ghz CPU. This is Corsair 2x1gb XMS Platinum @ 2.5-3-3-6 1T (highest overall timing is only 9). So maybe that's a DDR2/3 issue with higher clocks making the memory's latency rise.

 

[imperialreign]

IMO when it comes to ram speed is everything at a decent timing. I was speculating on a specific scenario were a user wasn't able to achieve DDR3-1600.

In the end, we will have to wait and see if they are able to actually get DDR3-1600 stable or not.

I tend to agree, but, in regards to DDR3, I don't think forcing lower CAS timings would prove to be that beneficial, even for users who can't push higher clock speeds.

In regards to my setup, I saw little benefit in forcing CAS6 versus my current timings at the same clock speed (1800). So little, actually, I decided that it wasn't worth the risk of instability to shave fractions of a ns off latency benches.

Granted, this is with a X38 chipset - AMD's might be a completely different ballpark in regards to actual operational performance of DDR3 on their setups.

 

[Polarman]

Pretty pointless.

Everyone wants a DDR2 Vs DDR3 comparison using the exact same settings to actually see it it really does improve at all.

 

[btarunr]

just my two cents...that looks more like gigabyte marketing than anything else...Gigabyte GPUs, Board, CPU Cooler (which cannot be qualified as an OC cooler by any means)

That cooler looks sufficient for a 2.60 to 3.10 GHz OC (voltage is constant), though the marketing crap is where they used two HD 2600 XT silent cards.

This p(r)eview shows performance scaling with an overclock that roughly sends the memory frequency to that of DDR3-1600.

 

[EastCoasthandle]

I tend to agree, but, in regards to DDR3, I don't think forcing lower CAS timings would prove to be that beneficial, even for users who can't push higher clock speeds.

In regards to my setup, I saw little benefit in forcing CAS6 versus my current timings at the same clock speed (1800). So little, actually, I decided that it wasn't worth the risk of instability to shave fractions of a ns off latency benches.

Granted, this is with a X38 chipset - AMD's might be a completely different ballpark in regards to actual operational performance of DDR3 on their setups.

Thus the reason why we have to wait until either
A. Someone tests it for themselves and post results
B. A review is made available regarding the subject

In the end, what can be said without speculating is that if a lower latency doesn't hurt performance and proves to be stable there is no reason why one shouldn't.

 

[Hayder_Master]

good , but the new gigabyte mobo support 1600 ram fsb , why they use 1333

 

[Bl4ck]

i know what my next cpu will be ;], time to go quad, Intel is to pricey for me , i rather give the money to AMD for now , not really need those "super pi breakers" ;]

 

[eidairaman1]

good , but the new gigabyte mobo support 1600 ram fsb , why they use 1333

next logical step, and 1600 ram is really for overclocking anyway.

 

[Hayder_Master]

next logical step, and 1600 ram is really for overclocking anyway.

i mean i want to see overclock the 1600 ddr3 , how much this mobo can handle

 

[nanohead]

I'm staying with DDR2 AM2+ until there is a material difference that I can feel in day to day work, or DDR3 is almost as cheap as DDR2 is now.

While the amount of data moved in and out of memory with DDR3 is somewhat greater, remember that DDR3 was really motivated by large power reductions (claimed 30%) over DDR2, and less so for lower latency.

And everyone complained with DDR2 CAS numbers went from 1.5 or 2 all the way up to 7 when DDR2 was initially released. Somehow we survived that horrific event in human history

 

[Lozza]

The overclocked chip scored a crunch-time of 46.613 s (1M).

That seems ridiculous. My Opteron 165 scores 30.907s in 1M at 2.8GHz so there is definitly something wrong with their testing, or the chips have somehow got worse.

 

[blueskynis]

Explanation

That seems ridiculous. My Opteron 165 scores 30.907s in 1M at 2.8GHz so there is definitly something wrong with their testing, or the chips have somehow got worse.

That's what I thought too. Is that a normal score?

EDIT: OK, here is the explanation:
http://www.xtremesystems.org/forums/showpost.php?p=3600818&postcount=3

I noticed a few mistakes/issues in their preview.First of all,their wPrime numbers are way off,it looks like wprime used only 1 core instead of 4 cores(take 46s for 3.12Ghz OCed AM3 Phenom II,apply the 3.89x scaling with 4 cores and you get right in the territory of known and valid scores for Phenom IIs at that clock).
Second,the OCed superPI number is not valid,it should be around 21.9s,instead they listed almost the same results for 1M for stock and OCed CPU.
Third,as they wrote the BIOS is not optimized for performance but for stability at this moment and they expect a performance lift(probably a few %s,still worth mentioning):

 

[eidairaman1]

That seems ridiculous. My Opteron 165 scores 30.907s in 1M at 2.8GHz so there is definitly something wrong with their testing, or the chips have somehow got worse.

DDR 1 vs DDR 3 dude, do the math, IIRC the HTT setup is a lowlatency solution, not a massive bandwidth solution like Core 2/P4/AM0D XP was.

also with this comparison

DDR1 PC 3200 (400MTS) VS DDR2 PC2 6400 (800MTS)

both get work done at the same time, but it takes the PC2 to run at a slower latency and higher bandwidth to get it done aka
222 5 (DDR) vs 444 10 (DDR2) etc now if DDR was Upped to PC 6400 it would eat PC2 6400 alive.

 

[spearman914]

500MHz is a fine oc. Keep it up, AMD!!

 

[Lozza]

DDR 1 vs DDR 3 dude, do the math, IIRC the HTT setup is a lowlatency solution, not a massive bandwidth solution like Core 2/P4/AM0D XP was.

also with this comparison

DDR1 PC 3200 (400MTS) VS DDR2 PC2 6400 (800MTS)

both get work done at the same time, but it takes the PC2 to run at a slower latency and higher bandwidth to get it done aka
222 5 (DDR) vs 444 10 (DDR2) etc now if DDR was Upped to PC 6400 it would eat PC2 6400 alive.

I see what you're trying to say, that now RAM is now more focused on bandwidth instead of latency.

But what I'm saying is that it should be able to produce better results despite this, because if you look at the price difference and how new the tech is you would expect it to be faster than a 2+ year old tech. It's the same as when DDR2 800 came out originally, the figures were similar to DDR 400 because of the latencies. The only thing going for it was the power consumption. That's only for the memory tests though.

But then if you look at Core 2 figures and Phenom figures in a superpi test the Core 2 is better even using the same RAM with the same latencies - so there is still scalability with respect to clock speed/computing power of the chip.

So I would expect the new chips to be better than the old one, even if they are using this "slower" memory.