Bulldozer running DDR3 2500 memory

[Pestilence]

Christ this makes me want to go ahead and order a Crosshair V

 

[sneekypeet]

Timings are poop, and WTF no numbers? What a waste of a screenshot

 

[Pestilence]

Timings are poop, and WTF no numbers? What a waste of a screenshot

You know how OBR works. Just enough to make us rant and rave.

 

[TheMailMan78]

Timings are poop, and WTF no numbers? What a waste of a screenshot

Maybe timings don't matter with BD?

 

[sneekypeet]

But if the scoring blows, what is the point of 2500mhz?

Even if AMD doesn't like tight timings, those timings suck!

 

[TheMailMan78]

But if the scoring blows, what is the point of 2500mhz?

Even if AMD doesn't like tight timings, those timings suck!

Who knows. Its all FUD until we get benches.

 

[btarunr]

Timings are poop, and WTF no numbers? What a waste of a screenshot

10-11-10-29T with CR1 is hardly poop for 2500 MHz. Something like 12-14-12-35 with CR2 would qualify as poop.

 

[Pestilence]

But if the scoring blows, what is the point of 2500mhz?

Even if AMD doesn't like tight timings, those timings suck!

Suck or not i'd still take them and that speed on my P67 rig

 

[sneekypeet]

10-11-10-29T with CR1 is hardly poop for 2500 MHz. Something like 12-14-12-35 with CR2 would qualify as poop.

Sorry when I ran 2500 a year ago, when it was tough to do and sticks weren't sold at 2500mhz, I was doing it with 9-9-9-27 on a set of Tridents. I still say those timings are poop, nobody wants cas10, the performance is lost at that point!

Already ram 2260 on my SB board, where it seemed to cap out, also at CAS9. I don't see any reason to go CAS10.

 

[cadaveca]

Suck or not i'd still take them and that speed on my P67 rig

P67 is very different, and has far better utilization than anything previous.

I doubt AMD is able to do the same. I understand why no numbers are shown, as performance is probably under NDA, but 5 GHz CPU and 2500 Mhz mem is nice and all...if it's on air. If it's under cold, then I care not, and I do think most people will feel the same.

If we cannot use it daily, this screenie is a waste of everyone's time.

 

[btarunr]

Sorry when I ran 2500 a year ago, when it was tough to do and sticks weren't sold at 2500mhz, I was doing it with 9-9-9-27 on a set of Tridents. I still say those timings are poop, nobody wants cas10, the performance is lost at that point!

But 2500 MHz with those timings is a rational increase considering memory with half the clock speed (1200 MHz with 5-5-5-12 CR1) is considered kosher or not-poop.

 

[cadaveca]

9-11-9-27 for 2400 MHz, makes the timings shown above so-so. Expected, but nothing earth shattering.

That's more about the rams than Bulldozer, anyway. Who care about timings? I wanna know NB speed..far more informative.

 

[sneekypeet]

But 2500 MHz with those timings is a rational increase considering memory with half the clock speed (1200 MHz with 5-5-5-12 CR1) is considered kosher or not-poop.

No offense BTA but we aren't the mass of idiots buying tech here. I personally don't want Kosher, and if you are going to brag about 2500mhz clocks, do it with "good" clocks and stop wasting time with this mediocre stuff.

Seriously I got to 2600mhz on Intel P55 before I needed cas10 for stability. I know they are completely different setups, but I still stand behind the fact that that speed at CAS10 is a kick in the grapes for benches, which seems to be his point.

EDIT: I digress, if this is an 8GB or 16GB kit of ram, I can't be so picky.

 

[TheMailMan78]

No offense BTA but we aren't the mass of idiots buying tech here.

Speak for yourself!

 

[erocker]

2500mhz on an AMD rig, regardless of timings is.. um.. neat? It means nothing though until we know how much bandwidth it's producing.

 

[TheMailMan78]

2500mhz on an AMD rig, regardless of timings is.. um.. neat? It means nothing though until we know how much bandwidth it's producing.

Or how much RAM it was running. For all we know it could have been 32 gigs.

 

[damric]

Better than my DDR3-1600 8-9-8-24 1T.

 

[seronx]

Timings are poop, and WTF no numbers? What a waste of a screenshot

10-11-10-29 are not bad timings

After 2000MHz timings are reversed

the actual timings do to the mem clock speed will be

8-8.8-8-23.2

Latencies don't matter past 2200MHz

 

[sneekypeet]

10-11-10-29 are not bad timings

After 2000MHz timings are reversed

the actual timings do to the clock speed will be

8-8.8-8-23.2

Latencies don't matter past 2200MHz

You have obviously never ran there then. I know all about latency and timings, and your explanation of these crap timings makes no sense.

 

[erocker]

10-11-10-29 are not bad timings

After 2000MHz timings are reversed

the actual timings do to the mem clock speed will be

8-8.8-8-23.2

Latencies don't matter past 2200MHz

You have obviously never ran there then. I know all about latency and timings, and your explanation of these crap timings makes no sense.

Running those sticks or not, we know nothing of the new AMD memory controller or how timings and frequency relates to performance on BD.

 

[seronx]

You have obviously never ran there then. I know all about latency and timings, and your explanation of these crap timings makes no sense.

2500 -> .8ns(2T) .4ns(1T)

.8 x (10-11-10-29)

8ns - 8.8ns - 8ns - 23.2ns

What is not to get?

 

[PaulieG]

But if the scoring blows, what is the point of 2500mhz?

Even if AMD doesn't like tight timings, those timings suck!

10-11-10-29T with CR1 is hardly poop for 2500 MHz. Something like 12-14-12-35 with CR2 would qualify as poop.

Sorry when I ran 2500 a year ago, when it was tough to do and sticks weren't sold at 2500mhz, I was doing it with 9-9-9-27 on a set of Tridents. I still say those timings are poop, nobody wants cas10, the performance is lost at that point!

Already ram 2260 on my SB board, where it seemed to cap out, also at CAS9. I don't see any reason to go CAS10.

I really think that whether the timings suck is determined by info we don't have. If this is just a 2x2GB kit, then yeah, it's kinda Meh. On the other hand if it's a 2x4GB set, then it's actually quite good, as I've yet to get any of my 8GB kits anywhere near 2500 no matter the timings. Way more stress with the IMC and Dimms with 4GB sticks.

 

[sneekypeet]

I really think that whether the timings suck is determined by info we don't have. If this is just a 2x2GB kit, then yeah, it sucks. On the other hand if it's a 2x4GB set, then it's not bad at all. Way more stress with the IMC and Dimms with 8GB.

I mentioned that at one point. I guess FUD is FUD until we really see the bandwidth results.

 

[PaulieG]

I mentioned that at one point. I guess FUD is FUD until we really see the bandwidth results.

Exactly.

 

[Damn_Smooth]

This is more crap from OBR??? Why do we keep discussing this guy? He is just trying to make a profit, and he probably doesn't even have an ES.

If he did have one, and bought it the way that he claims he did, he did not sign an NDA. So why hasn't he given us anything other than shopped superpi scores and crap that is useless to discuss?