• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

y-cruncher

sm625

sm625

Diamond Member
Have you seen the latest supposed bulldozer benchmarks?

There is this one: http://images.idgentertainment.de/images/idgwpgsgp/bdb/2248248/original.jpg

When I run y-cruncher on my E6600 3.2GHz, I get 41.5 seconds.

Settings are:
0- all in ram
0- Single-Threaded
1- 25,000,000 digits

This bulldozer core seems to run at the equivalent of a 4.3GHz conroe. That wouldn't be bad if it was a 2.8GHz chip but apparently this is a 4.2GHz chip. Is that an epic fail? Anyone know how a phenom II scores at 4.2GHz?
 
sm625 said:
Is that an epic fail?
Click to expand...

If I gave you $10,000 and your friend $60,000 and told both of you to go buy me the most awesome car you could get me for the money I gave you would it be "an epic fail" on your part if you brought me a 5yr old used Ford Focus and your friend brought me a spanking new Jaguar XF Premium Portfolio?

AMD has 1/6 the revenue of Intel, has 1/6 the resources to devote to R&D.

That they even come close, ever, to bringing to the market a semi-competitive product (and in case of bobcat, bringing a superior one to market) is winning when the odds are so overwhelmingly against them.

What's the epic fail in that?
 
Assuming the chip really does perform at these levels, I think the epic fail comes from the fact that a regular old athlon scores about the same. So why not just take 8 little athlon cores and string em together. Why make a behemoth 320mm2 monster if it is going to perform at the same level as 4x athlon II? 4x athlon II would be the same die size at 32nm and it would cost them virtually nothing to make it.
 
Until the chips are released for review no one knows. The last week or so the suppossed "reviews" keep getting wilder in predictions.
 
Last edited:
sm625 said:
This bulldozer core seems to run at the equivalent of a 4.3GHz conroe. That wouldn't be bad if it was a 2.8GHz chip but apparently this is a 4.2GHz chip. Is that an epic fail?
Click to expand...
Given the R&D resources and financial situation, AMD is doing all it can (and historically had delays, just cannot follow Intel on 2 year "tick-tock" route).

sm625 said:
Assuming the chip really does perform at these levels, I think the epic fail comes from the fact that a regular old athlon scores about the same. So why not just take 8 little athlon cores and string em together. Why make a behemoth 320mm2 monster if it is going to perform at the same level as 4x athlon II? 4x athlon II would be the same die size at 32nm and it would cost them virtually nothing to make it.
Click to expand...
That would result in a "Phenom II X8" without L3 cache. If add the L3 cache, then it would be a much bigger die. IMHO AMD began to realize that they cannot "out-core" Intel all the time (look at Sandy Bridge 4 cores versus Phenom II 6 cores). Calling BD (shared) "cores" as cores could be a marketing decision, IMHO should be called hardware "threads" (it may look embarassing if 8 BD "cores" are somewhat outperformed by 4 SB cores). 😉

richierich1212 said:
It's still too early to take early "benchmarks" as serious.
Click to expand...
Agree on that. These leaked benchmarks have to be taken with a grain of salt. The latest one, supposedly B2 stepping, has similar core performance with earlier B0 stepping (based on the SuperPi result) from a different source. However the "multi-core" performance has improved (4.60 vs 5.93 in Cinebench R11.5). Another 6 "core" B0 stepping for reference (simply extrapolate the score to 8 core), again from a different source. Somehow these does corroborate earlier rumors like this one.

Edit: Oops, forgot to compensate for clock speed differences (2.8Ghz/3.2GHz vs 3.6Ghz between B0 and B2 steppings). 😛

richierich1212 said:
Yeah that's what I saw as well. Just gotta wait another month for real reviews.
Click to expand...
Or wait for another "peformance/benchmark slip" (in the shades of Randy Allen's POV Ray demo). We already seen at least 3 slips ups (2 official, 1 unofficial). 😀
 
Last edited:
sm625 said:
Assuming the chip really does perform at these levels, I think the epic fail comes from the fact that a regular old athlon scores about the same. So why not just take 8 little athlon cores and string em together. Why make a behemoth 320mm2 monster if it is going to perform at the same level as 4x athlon II? 4x athlon II would be the same die size at 32nm and it would cost them virtually nothing to make it.
Click to expand...

Stranger things have happened. Intel went with Netburst and Willamette rather than multi-core P3 or Core (PPro) derivative at the time and it was a step back.
 
Idontcare said:
Stranger things have happened. Intel went with Netburst and Willamette rather than multi-core P3 or Core (PPro) derivative at the time and it was a step back.
Click to expand...
Again on the Intel front, from Northwood and Gallatin moving on to Prescott was also a step backwards (the Core Duo was the better replacement). Similarly on the AMD front, Athlon X2 from 90nm to 65nm with smaller L2 cache (from 1MB to 512KB, with essentially the same architecture). 🙂
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top