AMD Zambezi news, info, fans !

[informal]

I'm so disappointed now that I would now be ecstatic if Zambezi would have the same IPC as thuban in single thread integer code ,just the same (now it is -20 to -10% with wild fluctuations) . It doesn't need to have IPC improvements at all. Be the same as K10 in single thread code,have 33% more threads that would behave the same (or close) as 8 Thuban cores/threads in MT workloads and have a slight OC headroom Vs Thuban. This would be awesomely better than what we have seen with FX4100 and FX8150,it would be a whole other level of performance .

But... wishes are one thing and reality is another.I'm afraid that reality is going to be pretty bad for us enthusiasts and not as bad for the average Joe's who will see 8 unlocked cores at 3.6/4Ghz on the black box and usually buy it without much thinking.

[freeloader]

Guys,the reviewers already have all they need to test the platform correctly. AMD is not going to supply them with a last minute magic BIOS/OS workaround/whatever ,that is not how they operate. You cannot do that so close to product launch,it would invalidate all the work reviewers have done.And to make a decent review it takes a lot of time and testing.
This is it when it comes to performance. How is AMD going to spin this is unknown. I await for explanations why single thread IPC is not up like we were promised,why legacy SIMD(SSE) performance is slower than X6/X4,why multicore scaling is so poor in some workloads etc.

We won't get any explanations from AMD. This is a company that doesn't give a rat's ass about it's consumer base. They can't spin this in any positive manner as well. BD is a turd.

[dess]

The memory was set at 1866 CL8, and may be with Hypers even some more latency where lowered.

Isn't 20865 still too much?

Reading further that Romanian article I've found some other "interesting" bits (Googlish):

5066MHz with a voltage high frequency power was obtained 1.475va with a minimum of stability. I had full stability at frequencies of 4700-4800MHz, which is nothing like Sandy Bridge where most can run 5.1-5.2GHz stable.

Is it really the case...? I've heard most top at 4.7-4.8, just like BD. And some say you need a better cooler for that, just like with BD.

The stock voltage of 1.4v I noticed just scaling up to 1.475v, further having apparently need a better cooling. Worrying was the fact that although we have not increased too much pressure, ammeter began to show alarming levels when we made ​​some runs at 4.8 GHz of Cinebench ... 22.4 A, which translates to ~ 268W. Let us imagine for a second what it means and frequencies above 5 GHz 1.55v should be put into operation another reactor at Cernavoda.

How funny... Anyway, in case 1.55V is really too much (is it?), just why they didn't went up to 1.5? Perhaps because BD would have reached 5.1-5.2, indeed?

This is a company that doesn't give a rat's ass about it's consumer base.

BS! Then why they've made AM2+/AM3/AM3+ as much backwards/forwards compatible as possible? Why they've changed plans and now going to release Piledriver on AM3+, as well?

[MaddMutt]

Final silicon and hardware probably. That doesn't mean everything is final.

If it's slower than K10.5, they shouldn't even bother to release it. Simple as that.

I have the same questions????????

If it is slower than the K10.5 then just shrink the old arch and let it run until you fix your NEW MONEY MAKER!!!!!!

What is the point reading/hearing about the new AMD BD, when all you hear is.............

Here is our new product, The AMD FX-xxxx BLAH....BLAH....BLAH.....It's the first TRUE 8 CORE desktop processor .....BLAH.......BLAH.....BLAH.....AMD HAS SET A NEW WORLD RECORD AND IS THE FIRST TO RELEASE A 10GHZ desktop processor BLAH.......BLAH......BLAH.......AS YOU CAN SEE FROM THE BENCHMARKS OUR NEW PROCESSOR IS ONLY 2% SLOWER THAN A 286

[LightSpeed]

If true, a sad way to bring back the FX name.

Im still hoping for something magical, but that hope is very, very slim now.

[jabway]

If this is the actual performance of BD(and it looks like it is), this is one of the worst product launches ever, even worse than the Phenom I launch.

[Opteron146]

Can I sincerely ask how are these (i7 numbers) possible with stock ram frequency? I mean, top bandwidth for DDR3-1600 (the max. that SB officially supports) 12800 MB/s... Even with DDR3-1866 you get 14933...
http://en.wikipedia.org/wiki/DDR3

Ehhmm ... you heard of dual channel my friend?
Take your numbers times 2, then you have the theoretical maximum bandwidth ... intel is quite good in using it, AMD well - at least better then before and they have lots of cache this time. Memory should not be an issue as long as the caches are fine and fast ;-)
However, it seems they are not

[AAbenson]

actually this is so bad that im in doubt that its true

[Manicdan]

Reading further that Romanian article I've found some other "interesting" bits (Googlish):

Is it really the case...? I've heard most top at 4.7-4.8, just like BD. And some say you need a better cooler for that, just like with BD.


How funny... Anyway, in case 1.55V is really too much (is it?), just why they didn't went up to 1.5? Perhaps because BD would have reached 5.1-5.2, indeed?

im thinking the reviewer needs to learn a little more about overclocking. using the stock heatsink is a bad idea when it comes to a chip that clearly scales due to cold, not volts. SB is rather simple, you up the volts a little and just overclock, ignoring temps. but with thuban and apparently the same with BD, you need to make sure you balance temps with volts. seeing 4.8ghz on stock heatsink is quite nice, since it does confirm that 5ghz should be easy for WC or bigger heatsinks.

[zalbard]

seeing 4.8ghz on stock heatsink is quite nice, since it does confirm that 5ghz should be easy for WC or bigger heatsinks.

Not necessarily.
See Gulftown, for example. Air and water OCs do not differ significantly (unless you're pushing really high voltages), and it also scales well with LN2.

[Manicdan]

Not necessarily.
See Gulftown, for example. Air and water OCs do not differ significantly (unless you're pushing really high voltages), and it also scales well with LN2.

i dont look at just LN2 when i say scales with cold. when they were showing off the 8.4ghz record, they had speeds for all forms of cooling and it got faster and faster with every increase in cooling power.

[xdan]

Yeah, sure, Monstru & Matose, team lab501, MOA 2011 and GOOC 2010 winners, Hwbot Country Cup two times winers , should learn about overclocking.
They didn't sad that on overclocking they used BOX cooler(by the way how that is not a retail if they have the cooler)- by the way may be the box cooler is a WC.
And in ROMANIA it's pretty cold now, 14-15C, in the morning 7-8C. It's easy to open the window and bring some cold.

[Sn0wm@n]

chew* seems quiet ....


ill wait to see what he has to show us once nda drops and then ill make my judgment

[muziqaz]

Funny how I get flamed when I say that most of the BD threads have loads of crap in it
If we take 501 test results and put them against what chew and Movieman was talking about and how they were talking about BD, it doesn't make any sense whatsoever. I am not saying 501 results are fake, but clearly something is wrong with that system. I doubt that Movieman and chew would be talking so positively about BD if it was performing like 501 tests show.
For me, if BD is between 2500 and 2600(closer to 2600) in performance, I will buy it next weekend or whenever it shows up in that 3rd world country they call UK, but if BD performs like it is shown on 501 site, then I will just grab 2600K and be happy with it, or just wait a bit longer for SB-E. No need to write pages and pages of rants on how AMD failed releasing BD. Just buy whatever you think will fit your needs

[skill123]

Problems , again and agian ; now i have to decide what to chose asrock fatality P67 + i7 2600k or Crosshair V + bulldozer 8170.
Every new CPU design has problems and needs more tweaking but for us end users , we need a good CPU now , not 1 year from now .
I'll wait the release of bulldozer then i'll make my mind. ( i realy hope amd cpu to be better and to go with amd , but it wont matter that much if it sucks that hard , so i have no problem going with intel).

[mAJORD]

not being much faster than the previous flagship isn't really such a bad thing. The problem I have though is Thuban has been out now for 18months, and has recieved 1 speed bump almost a YEAR ago now. so it's now very very far behind the game.. That's why I expect better than this.. time waits for no one.

AMD architecture launches have always been pretty flakey. So I hold some hope for the future. What I question though, is how higher frequency can this architecture scale to? and what IS its longterm target?

I expected a longer pipeline, higher frequency design than K10, but this doesn't seem right. (and still no word on official pipeline length?)

[Negaterium]

Poor AMD. We had hopes... Looks like waiting for Ivy Bridge is all that's left.

[muziqaz]

I remember when I upgraded from athlon x2 5000@3.1ghz to phenom 9600@2.3ghz. I could not OC phenom as I had crappy motherboard with some crappy ram. And everyone considered K10 launch a disaster. BUT that phenom was pulling better or equal numbers in all the apps I was using which where single or dual threaded, and obliterated 3.1ghz athlon in multithreaded apps. So even so called disastrous CPU like Phenom was quite a progress for AMD.
So according to 501 numbers BD is even worse than k10 launch

[mAJORD]

I remember when I upgraded from athlon x2 5000@3.1ghz to phenom 9600@2.3ghz. I could not OC phenom as I had crappy motherboard with some crappy ram. And everyone considered K10 launch a disaster. BUT that phenom was pulling better or equal numbers in all the apps I was using which where single or dual threaded, and obliterated 3.1ghz athlon in multithreaded apps. So even so called disastrous CPU like Phenom was quite a progress for AMD.
So according to 501 numbers BD is even worse than k10 launch

Yes, phenom had lower single threaded performance than K8 at launch.. by quite a bit actually ( 2.3Ghz, vs 3.2Ghz ! .. the IPC adv was never gonna do much for that! ). But as you say, it had 4 cores, and made up for it since even then, quite a few apps in reviewers benchmark suites scaled to use 4 threads.

Still, was quite unappealing though. A lot of us still had little use for 4 cores in reality, hence I never upgraded my 6400+ in my daily machine until Phenom II came along. All my Phenom CPU's were for testing purposes only.

[informal]

I remember when I upgraded from athlon x2 5000@3.1ghz to phenom 9600@2.3ghz. I could not OC phenom as I had crappy motherboard with some crappy ram. And everyone considered K10 launch a disaster. BUT that phenom was pulling better or equal numbers in all the apps I was using which where single or dual threaded, and obliterated 3.1ghz athlon in multithreaded apps. So even so called disastrous CPU like Phenom was quite a progress for AMD.
So according to 501 numbers BD is even worse than k10 launch

Yes,you can call Agena a pretty good successor to successful K8.2x more cores,integer IPC was 10-15% better,in vector SIMD it was 50-70% faster due to 2x wider SSE pipelines.Only bad thing was poor clocking ability and power draw. After Agena we had excellent Deneb and then even better Thuban which brought us 50% more cores within same TDP(!),similar clocks,Turbo Core(!) and great OC headroom.All on same 45nm as Deneb,pretty impressive IMO.

Now we have around 15-20% worse IPC vs Thuban/Deneb(which practically negates the 33% more cores uplift we have on paper),same power draw,great OC headroom with high power raw (which is understandable). So compared to even original K10,this thing is actually worse.

[muziqaz]

Yes, phenom had lower single threaded performance than K8 at launch.. by quite a bit actually ( 2.3Ghz, vs 3.2Ghz ! .. the IPC adv was never gonna do much for that! ). But as you say, it had 4 cores, and made up for it since even then, quite a few apps in reviewers benchmark suites scaled to use 4 threads.

Still, was quite unappealing though. A lot of us still had little use for 4 cores in reality, hence I never upgraded my 6400+ in my daily machine until Phenom II came along. All my Phenom CPU's were for testing purposes only.

As I said, my agena was faster than my athlon in single threaded apps same as in dual threaded apps. And it had 800mhz disadvantage than my k8. Now take 3.2ghz Thuban and put it against 2.4ghz BD, man that would be a straight up slaughter. Even clock for clock it would be murder all over the place. The only thing is BD can clock higher, but didn't we have P4 which was relying on clock speeds instead of strong core?

[mAJORD]

mm maybe faster than your athlon, in some ST benchmarks, but i'm comparing top bin vs top bin.. The 6400+ was faster than 9600 without fail at 1-2 threads, even with its 128bit SSE capabilty

[vietthanhpro]

11/10 is Patch Tuesday

[undone]

11/10 is Patch Tuesday

corrected. Nov.10 is Thursday, so it's Oct.11. I was aware that this forum rule date in mm/dd/yyyy.

edit: If i remember correctly long time no see vietthanhpro, where have you been

[muziqaz]

mm maybe faster than your athlon, in some ST benchmarks, but i'm comparing top bin vs top bin.. The 6400+ was faster than 9600 without fail at 1-2 threads, even with its 128bit SSE capabilty

Yes but again agena had low clocks. BD on the other hand has higher clocks than previous gen.