• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Bulldozer Finally!

Last edited:
Of course it was ES. There's no if. Besides, they used 1333Mhz on a platform where the native speed for the IMC is 1866Mhz :rolleyes: Do people test Sandy Bridge at 800Mhz instead of 1333Mhz?
 
theres maybe another issue or will run slightly slower on am3 with BD, the System Bus(MT/s) on the am3+/BD is at 5200, the System Bus(MT/s) on the am3/phenom II is 4000.

i know it maybe a very very small performance hit
 
Last edited:
gareth170 said:
theres maybe another issue or will run slightly slower on am3 with BD, the System Bus(MT/s) on the am3+/BD is at 5200, the System Bus(MT/s) on the am3/phenom II is 4000.

i know it maybe a very very small performance hit
Click to expand...

PII doesn't come close to saturating the bandwidth. It's likely BD may only just come close, or may slightly go over.
Can't wait to find out.
I am heavily debating switching, but knowing I'll see a decrease in the games I play, but the BD will last longer.
 
Martini1991 said:
PII doesn't come close to saturating the bandwidth. It's likely BD may only just come close, or may slightly go over.
Can't wait to find out.
I am heavily debating switching, but knowing I'll see a decrease in the games I play, but the BD will last longer.
Click to expand...
PII is 2000 x2 = 4000, BD is 2600 x2 = 5200

i think it goes by the HT
 
Martini1991 said:
PII doesn't come close to saturating the bandwidth. It's likely BD may only just come close, or may slightly go over.
Can't wait to find out.
I am heavily debating switching, but knowing I'll see a decrease in the games I play, but the BD will last longer.
Click to expand...

Saturating is not always the only factor, even with small bits of data latency can be reduced with higher frequencies.
 
Martini1991 said:
But you can run a HT over 2.6GHZ on a Phenom II.
Click to expand...
i'm talking about stock, which would be at it's max..

yes you can run HT over 2.6GHZ but you need to up the bus speed to do that

with amd+ HT 3.1 it may/should allow a higher range without the need to up the bus speed
 
Last edited:
gareth170 said:
i'm talking about stock, which would be at it's max..

yes you can run HT over 2.6GHZ but you need to up the bus speed to do that

with amd+ HT 3.1 it may/should allow a higher range without the need to up the bus speed
Click to expand...

You never had to raise the bus to increase the HTL.
It just had to be either the same as, or less than the CPU NB.

So, on Zambezi, stock CPU NB is 2.6GHZ and HTL of 2.6GHZ.

The highest I took my Thuban was 3.2GHZ CPU NB.

There might be more headroom with the CPU NB, but the HTL never affected performance. The higher the HTL frequency, just allows more bandwidth, but it never got saturated even at 2GHZ when running a massively overclocked Thuban.
You reported previously that you noticed no difference upping the HTL.
You could try it yourself, run your HTL at 2GHZ running a benchmark at 4GHZ, then up the HTL to 2.6GHZ and run the benchmark again.

Latency you say fina8ly? Explain that to me, I know what latency is, but I have no idea how to apply it here.

I've not seen any latency figures of either of the HT standards.
 
Martini1991 said:
I am heavily debating switching, but knowing I'll see a decrease in the games I play, but the BD will last longer.
Click to expand...

But if your not encoding or anything and just gaming why take the FPS drop? it may not be that noticeable either but why spend the cash when you could just wait til next year and see what IVY does, which by all likelihood outperform in your games and also last even longer, just a thought.

Heck there might even be some of the newer BD chips out next year that will suit you better if thats the route you wanna go
 
Martini1991 said:
You never had to raise the bus to increase the HTL.
It just had to be either the same as, or less than the CPU NB.

So, on Zambezi, stock CPU NB is 2.6GHZ and HTL of 2.6GHZ.

The highest I took my Thuban was 3.2GHZ CPU NB.

There might be more headroom with the CPU NB, but the HTL never affected performance. The higher the HTL frequency, just allows more bandwidth, but it never got saturated even at 2GHZ when running a massively overclocked Thuban.
You reported previously that you noticed no difference upping the HTL.
You could try it yourself, run your HTL at 2GHZ running a benchmark at 4GHZ, then up the HTL to 2.6GHZ and run the benchmark again.

Latency you say fina8ly? Explain that to me, I know what latency is, but I have no idea how to apply it here.

I've not seen any latency figures of either of the HT standards.
Click to expand...

Anything that has a frequency has a latency but it will not always have a noticeable effect because of other bottlenecks and hold ups.

Example: A program uses 1 in 10 cycles of CPU at 1Ghz so that CPU is not saturated at all, even though that program was using little of the CPU, increasing to 2Ghz & still using 1 in 10 cycles is more performance less latency, but because of possible other dependences, at 2Ghz it may only use 1 in 20 cycles so you get no gain but in both cases the CPU was not saturated .
 
Last edited:
Martini1991 said:
You never had to raise the bus to increase the HTL.
It just had to be either the same as, or less than the CPU NB.

So, on Zambezi, stock CPU NB is 2.6GHZ and HTL of 2.6GHZ.

The highest I took my Thuban was 3.2GHZ CPU NB.

There might be more headroom with the CPU NB, but the HTL never affected performance. The higher the HTL frequency, just allows more bandwidth, but it never got saturated even at 2GHZ when running a massively overclocked Thuban.
You reported previously that you noticed no difference upping the HTL.
You could try it yourself, run your HTL at 2GHZ running a benchmark at 4GHZ, then up the HTL to 2.6GHZ and run the benchmark again.
Click to expand...
is System Bus(MT/s) meaning cpu-nb or ht?

ether way on am3/PII the max on both are 2.6ghz using just the multipliers anything over that would need you to up the bus speed/fsb.
 
gareth170 said:
is System Bus(MT/s) meaning cpu-nb or ht?

ether way on am3/PII the max on both are 2.6ghz using just the multipliers anything over that would need you to up the bus speed/fsb.
Click to expand...

The max on CPU NB multiplier wasn't 13 was it?
I don't remember, my Thuban was multi 10 chip.

Although, if the limit is 13, a bit of a coincidence is it not? ;)
 
Martini1991 said:
The max on CPU NB multiplier wasn't 13 was it?
I don't remember, my Thuban was multi 10 chip.

Although, if the limit is 13, a bit of a coincidence is it not? ;)
Click to expand...
yes both CPU NB and HT

thats what im getting at, using BD in a am3 board you've got no headroom on the multipliers, and at stock it may default to 2.0ghz on both.
 
Last edited:
Bleek said:
Is it?

Can you post the benchmarks of the retail CPU you've based this statement off. I must have missed them? :confused:
Click to expand...

think he was looking at the benchmarks i found over at xs forums from sisandra here;

http://webcache.googleusercontent.c...zer&cd=2&hl=de&ct=clnk&gl=de&client=firefox-a

Although take it with a large pinch of salt as they are supposed to have been done on an old engineering sample with the prefetcher disabled which would completely destroy the results, sandy bridges advancements have mainly been from the prefetcher improvements apparently.
 
Martini1991 said:
Who has 1866MHZ RAM though? Not many people at all...
Click to expand...

Me :) ;) :) - ready and waiting dude lol

We've been through this debate of HT 3 v's HT3.1 at XS Forums - BD will NOT be able to saturate the current HT3 limits let alone the new one. If that was the case we would not have certain AM3+ mainboards based on the 890 Chipset ;)
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom