if the chips arent ready for production, or if they simply dont have the manufacturing capacity at the moment due to other commitments then it would seem daft to go releasing any actual information as to the performance of the chip as its just giving your rivals time to either change there plans and bring forward a more powerfull release of there own, or adjust prices of existing stuff to sway over anyone planning to wait for it. Better to keep everyone guessing imo and spend the time making sure the chips a good one.
-
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!
- Thread starter AceTK
- Start date
Caporegime
- Joined
- 18 Sep 2009
- Posts
- 30,294
- Location
- Dormanstown.
September.
Soldato
- Joined
- 31 May 2009
- Posts
- 21,457
Thx, guess they'll have to be good, giving Sandybridge a good 6 months headstart.
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,342
- Location
- Planet Earth
Corrected for you.
Considering that I posted almost all of the recent Bulldozer leaks/info first in this thread too some of us have an interest in both processors. Perhaps you missed the part the part about Trinity being Bulldozer based. Llano is also based on the same 32NM process as Bulldozer and is the first to use the second generation Turbo Core Bulldozer will use. On top of this it does seem the shared memory controller is not affecting performance too much which is a good sign for Trinity. Considering that AM3+ Bulldozer processors will be priced at the same level as Core i5 processors,Llano and Trinity performance will be relevant for many people. Llano will also be found in laptops too unlike Bulldozer ATM.
Last edited:
Will Llano cpu's be on sale on 14th june? If not will there be reviews out using the final stepping on 14 june?
I dont quite get this part now then, as looking at the results (unless of course they are faked) then what you are saying is when the guy clocks it up to 4ghz and still getting disappointing performance, it should be indicative of real world performance as the only difference is clock speed?
Surely there must be more to it than that or have we been to overly hopeful of BD? Or have I taken your words out of context, which of course is totally plausible too.
Llano does seem pretty awesome for both laptops and low-end desktops (office PCs or ones just used for browsing, Office, etc.). You could probably get a motherboard/CPU/IGP/RAM combo for £150-160. Throw in a 500 GB HDD, half decent PSU and case and you've got an awesome machine for £250.
As for Zambezi, I'm rapidly losing interest. If I do manage to somehow get a new PC for work, it'll have to have great single-threaded performance and I just don't think I'm going to get that from Zambezi. Even if I did, I doubt it'd be released in time anyway.
As for Zambezi, I'm rapidly losing interest. If I do manage to somehow get a new PC for work, it'll have to have great single-threaded performance and I just don't think I'm going to get that from Zambezi. Even if I did, I doubt it'd be released in time anyway.
Last edited:
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,342
- Location
- Planet Earth
http://forums.overclockers.co.uk/showpost.php?p=19341142&postcount=1990
The video states Llano will be launched in the middle of this month.
What I am saying is that a.) I have no idea if the results are real or not and that b.) if the results are real, then there is an extremely high probability that the results will not reflect actual performance.
I don't even need to look at the results to tell you that. To date, I have seen nothing (high or low) that I believe actually reflects actual performance. I have teams of benchmarking people in our own labs that are running benchmarks today, on the most up to date silicon, and they still have a way to go with tweaking to get the best performance. What makes you believe that some person off in their basement somewhere has a better read on the final performance?
Most people here know this, but these 'leaked' engineering samples are the only information we have to go by at the moment.
If they're not indicative of final performance, then AMD should really release some more accurate information to prove so.
Oh I don't believe its final performance at all and thats what I was stating in the post you originally quoted. It was the part where you said ES were just lower clocks with nothing else disabled (Well thats what I took it to mean). I thought there would have been more things not right or intentionally prevented from functioning.
But its OK saying you have teams of benchmarking people in your own labs, when the end-user and people who enjoy discussing and speculating on things see no official things out there were gonna speculate on whatever is to hand. Unfortunately speculation can sometimes stick in peoples minds, as like I said I found it surprising that people were taking the results of this benchmark as an indication of final performance, as that performance would make no sense at all.
PS. Thought you were on your hols, you should be drinking whatever the locals have to offer not on here answering our speculations
I'd say there's a pretty high probability of this guy's theory accurately calculating current B0 engineering sample performance at proper clocks / TDPs.
The numbers he calculates are within such a small margin of error to what they *should* be if the theory is correct that I'm pretty certain he's got it sussed.
If this is indeed the case, then it's very good news indeed. Even at 3.2Ghz, the 8 core will be an absolute beast. Smoking a 980x in Cinebench.
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,342
- Location
- Planet Earth
It seems there was a bug with Cool'n'Quiet which meant that some of the recent Llano samples ran at half their rated speed. With it is switched off the CPU benchmark scores are much higher:
http://translate.google.com/transla...www.coolaler.com/showthread.php?t=266878&pg=5
Supposedly,the CPU was overclocked to 5.4GHZ although the scores do look a tad low for such a large overclock. It could be quite possible that there is some throttling happening.
http://translate.google.com/transla...www.coolaler.com/showthread.php?t=266878&pg=5
Supposedly,the CPU was overclocked to 5.4GHZ although the scores do look a tad low for such a large overclock. It could be quite possible that there is some throttling happening.
Have a look @ CPUz its only running one core one thread.
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,342
- Location
- Planet Earth
Good spot!
this is INSANE! 8 cores at 5.4ghz theoretical max - that is a lot of folding potential
i'd love if the underdog knocked off intel at a fraction of the cost - that would shake things up a bit
this is INSANE! 8 cores at 5.4ghz theoretical max - that is a lot of folding potential
i'd love if the underdog knocked off intel at a fraction of the cost - that would shake things up a bit
5.4ghz was on a llano sample, which for a start is a different architecture to the bulldozer chips, includes a graphics unit, and because it wasn't an official announcement, no reliable.
8 cores at 5ghz+ though would pretty damn cool
