• 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.

AMD FX-8120P benchmark from Coolaler

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
BlueBlazer said:
If you insist on that, then can we expect more delays (because silicon re-spin takes time, in months)? Also I happen to mention in another thread about lack of any announcement on shipping dates, as well as launch dates. Hopefully it will not be another last minute "60-90 days" as shown by AMD back in June. :hmm:
Click to expand...

Respin from early Jun to early Sept. 1 month to ramp, 2 weeks to ship, that means around the middle of October for launch. The NDA in the competition AMD held, where the prize was BD, lasts till Oct 12 iirc.
 
BlueBlazer said:
That reference was based on fps with multi-threaded video transcoding. Its not based on trancoding time, such as this one here. :hmm:

Huh? The article says the AMD FX processor result is 223 fps, then why do you need to multiply the result by 1.17 to inflate it to 260? Where does that 1.17 come from? I'm using Core i5 result of 188 fps from that article as a base.. :hmm:
Click to expand...

My results come from fps based results where 2500 scored 257fps and 2600 263fps.

The 17% is the multithreaded SSE performance difference between the 8120 and the 8150.
 
Arzachel said:
Respin from early Jun to early Sept. 1 month to ramp, 2 weeks to ship, that means around the middle of October for launch. The NDA in the competition AMD held, where the prize was BD, lasts till Oct 12 iirc.
Click to expand...
That doesn't leave space for lots of revisions. I can say optimistically at most 3 revisions (tight back-to-back schedules) while realistically only at most 2 revisions (from B2F stepping to B2G stepping?) considering debugging/testing/validation cycles at motherboard manufacturers. :hmm:
 
Riek said:
My results come from fps based results where 2500 scored 257fps and 2600 263fps.
Click to expand...
Source of this result? This one sounds more like single threaded because Core i7 2600K clock speed is slightly higher than Core i5 2500K. Thus the slight improvement only. Toms Hardware results (if you take the transcoding time and convert to performance ratio) is even higher than Guru3D's (faster by 100%)! :hmm:

Riek said:
The 17% is the multithreaded SSE performance difference between the 8120 and the 8150.
Click to expand...
Source of this information? The article did not state which processor model by the way, just "AMD FX". :hmm:
 
BlueBlazer said:
Source of this result? This one sounds more like single threaded because Core i7 2600K clock speed is slightly higher than Core i5 2500K. Thus the slight improvement only. Toms Hardware results (if you take the transcoding time and convert to performance ratio) is even higher than Guru3D's (faster by 100&#37😉! :hmm:

Source of this information? The article did not state which processor model by the way, just "AMD FX". :hmm:
Click to expand...

http://www.servethehome.com/intel-xeon-e3-1220-sandy-bridge-benchmarks-review/




http://www.tomshardware.co.uk/core-i7-3960x-x79-performance,review-32272-11.html
You talk about these time based toms hardware results? Where a 6core/12T SB-E is 17% faster than the i7 2600K with 50% more cores?

And where the 2600K matches the 990x version? by ~10%.
Click to expand...



Source of this information? The article did not state which processor model by the way, just "AMD FX". :hmm:
Click to expand...
They mentionned equally priced FX 8core to an i5. And unless the FX8150 rumoured priced is 60$ less that cpu would be the FX8120 (3.1Ghz vs 3.6Ghz).
 
Last edited:
BlueBlazer said:
Source of this information? The article did not state which processor model by the way, just "AMD FX". :hmm:
Click to expand...

First they showed a comparison between a new unspecified AMD FX processor and an unspecified Intel processor i5 Sandy Bridge, with the help of the program Handbrake a video of 5 minutes is converted to H.264 video in SD resolution. The AMD FX processor with eight cores perform this function with an average of 223 frames per second, the i5 with four cores came in at 188 fps. Both systems will be comparable in price according to AMD, which it wants to show that AMD a better price / performance offering.
Click to expand...
....
 
Riek said:
http://www.servethehome.com/intel-xeon-e3-1220-sandy-bridge-benchmarks-review/
Click to expand...
Looks as borked as another benchmark I saw (which Phenom II X6 should be faster than Core i5 2500K). Its possible its limited to 4 threads or cores not fully utilized (as in this example). :hmm:

Riek said:
They mentionned equally priced FX 8core to an i5. And unless the FX8150 rumoured priced is 60$ less that cpu would be the FX8120 (3.1Ghz vs 3.6Ghz).
Click to expand...
IMHO its not the FX-8120 at all. Equally priced? See below....

Abwx said:
....
Click to expand...
I have also noted that "Both systems will be comparable in price". That is if you also take into account the (cheaper) price of AMD boards. :hmm:
 
BlueBlazer said:
I have also noted that "Both systems will be comparable in price". That is if you also take into account the (cheaper) price of AMD boards. :hmm:
Click to expand...

Probably that it s either a 6XXX or a 8120 , since the price should
be comparable to a 2500K , MBs doesnt make a lot of difference
since it s the low cost 1155 against not so cheap AM3+....
 
Abwx said:
Probably that it s either a 6XXX or a 8120 , since the price should
be comparable to a 2500K , MBs doesnt make a lot of difference
since it s the low cost 1155 against not so cheap AM3+....
Click to expand...
If you are going head-to-head against competitor CPU, then wouldn't you use the best CPU to showcase? Why would AMD give away low numbers with a lesser CPU? Also do most demonstration machines use low end boards? For your information, most demonstration machines always use high end boards, and you can find this trend throughout the other demonstrations (by both camps). :hmm:
 
BlueBlazer said:
Looks as borked as another benchmark I saw (which Phenom II X6 should be faster than Core i5 2500K). Its possible its limited to 4 threads or cores not fully utilized (as in this example). :hmm:

IMHO its not the FX-8120 at all. Equally priced? See below....

I have also noted that "Both systems will be comparable in price". That is if you also take into account the (cheaper) price of AMD boards. :hmm:
Click to expand...

lol how contradictory can you get?

FX8120 is priced above the 2500.... yet you say it isn't that cpu because you have to take the system price into account and that AMD boards are cheaper.

So what do you get when you have slightly more expensive + cheaper? seems to me that overal system price would have a lower difference between it and that i5 wouldn't it?


Like i already mentionned in another rambling, you cannot compare those results! it makes a huge difference which you select, how big it is etc.. Funny you dismiss results i bring up as wrong because it doesn't prove your point.... at least i'm big enough to admit that you cant draw any conclusions from the known results. Unlike what I initially thought due to those performance numbers.

I also pointed you to a test where a 6core/12T Sb-E performs only 17% better in handbrake then the 2600. so either scaling becomes alot less when you add more cores or their is a limit to the number of threads.... so you cannot deduct performance simply by calculating the core difference. so again the conclusion is that you can't draw any conclusions on those numbers using differences from the same program with other tests.
 
Last edited:
Riek said:
lol how contradictory can you get?

FX8120 is priced above the 2500.... yet you say it isn't that cpu because you have to take the system price into account and that AMD boards are cheaper....

So what do you get when you have slightly more expensive + cheaper? seems to me that overal system price would have a lower difference between it and that i5 wouldn't it?
Click to expand...
Which begs to question.... Are high end boards for AMD usually cheaper than high end boards for LGA1155 platform? When I mentioned cheaper, I do mean across the board. Check the price of ASUS ROG Crosshair V Formula board against LGA1155 boards such as GigaByte G1.Sniper2... :hmm:

Riek said:
Like i already mentionned in another rambling, you cannot compare those results! it makes a huge difference which you select, how big it is etc... Funny you dismiss results i bring up as wrong because it doesn't prove your point.... at least i'm big enough to admit that you cant draw any conclusions from the known results. Unlike what I initially thought due to those performance numbers.

I also pointed you to a test where a 6core/12T Sb-E performs only 17% better in handbrake then the 2600.... so either scaling becomes alot less when you add more cores or their is a limit to the number of threads.... so you cannot deduct performance simply by calculating the core difference. so again the conclusion is that you can't draw any conclusions on those numbers using differences from the same program with other tests.
Click to expand...
My answer here.....
If you take those results as scaling by cores/threads (just look at the delta between the 4-core Core i5 and 8-"core" AMD FX in that article), then you have to find a comparable benchmark results which scales with cores/threads. Thus I picked the Guru3D's as reference as it shows this characteristic. The other Handbrake results looks borked, IMHO (including the Toms Hardware one). :hmm:
Click to expand...
 
I'm sure even the graphics drivers will have to be rewritten extensively to optimize for BD. AMD should logically be first out of the gates with optimized drivers. From those before and after results we should get an idea of what improvements others will get once they optimize and recompile.
 
Riek said:
lol how contradictory can you get?

FX8120 is priced above the 2500.... yet you say it isn't that cpu because you have to take the system price into account and that AMD boards are cheaper.

So what do you get when you have slightly more expensive + cheaper? seems to me that overal system price would have a lower difference between it and that i5 wouldn't it?


Like i already mentionned in another rambling, you cannot compare those results! it makes a huge difference which you select, how big it is etc.. Funny you dismiss results i bring up as wrong because it doesn't prove your point.... at least i'm big enough to admit that you cant draw any conclusions from the known results. Unlike what I initially thought due to those performance numbers.

I also pointed you to a test where a 6core/12T Sb-E performs only 17% better in handbrake then the 2600. so either scaling becomes alot less when you add more cores or their is a limit to the number of threads.... so you cannot deduct performance simply by calculating the core difference. so again the conclusion is that you can't draw any conclusions on those numbers using differences from the same program with other tests.
Click to expand...

$230 vs $220. A meager $10 difference...

And no, you can't compare the vanilla 2500 with it. It's not unlocked, while the 2500K and FX-8120 are.
 
I seen who presented the benchies. But these are not offical benchmarks . SO wait for them . I never ever thought an enthusiast board would come down to debating price performance on a second tier product. IF a $100 is all that . I sugjest getting a new life or a forum that isn't made up of enthusiast. IT really doesn't matter how BD performs it will still be fast and if thats what people want good for them . Intel isn't going to raise or lower pricies based on AMD. Out of all the news I have read the last 2 weeks the problem with intels 22nm tri-gate yields was the most news worthy and troubling for those of us hoping for a 22nm price drop.
 
LOL_Wut_Axel said:
$230 vs $220. A meager $10 difference...

And no, you can't compare the vanilla 2500 with it. It's not unlocked, while the 2500K and FX-8120 are.
Click to expand...
Are we even sure about 8120's price? All we have seen are early price details. I wonder if price will actually go down shortly after launch.
 
Bd has had the telltale signs of being a disappointment for some time.
1) The sacking of the ceo as soon as sb was released
2) Delay after delay after delay
3) The refusal of AMD to throw the media/public a bone in regards to previews or benchmarks (this can partly be attributed to #2) citing Whatitscalled's Law saying they were worried about stalling sales of their current line up if people knew the coming gen was such a leap in performance they would be stuck with a backlog of old unsellable cpus. Now it seems they would kill to be in that situation because that would mean they had a killer chip on their hands. The truth is probably if people knew how Bd is to perform months ago they wpuldve went Intel. now there's a lot of people stuck with am3 boards pretty much committed to Bd.

Now I hear some people saying "even if Bd isn't that great, the refresh will be able to improve upon the design and have much better performance etc"

If you have waited through the numerous delays, refused to buy sb in the hopes Bd would
Match it, and now believe that the next BD will be on time, let alone be able to compete with Ivy Bridge with Haswell closely following, please contact me asap, I have some magic beans I Want to sell you.
 
Last edited:
I don't know how people can draw performance conclusions on seriously context/fact lacking leaks and anecdotes. AMD stated that the FX desktop processors would ship before the end of September in prep for an October launch. We are going to have to wait until the NDA drops to get any good info.

Another thing to consider is that BD is not targeted primarily as a desktop architecture. AMD designed it as a server chip that would do well in the multithreading, cloud, and virtualization workloads. It would also be die size efficient and have robust energy saving controls. From what I have seen data center and supercomputing clients are optimistic about what they are going to be able to do with these chips. They drop into existing platforms and are going to trounce AMD's previous chips. The margins in the server fields are very attractive compared to the desktop market. This chip is designed to get AMD's market share back in the server market. The desktop market is secondary to that consideration.

Secondly, AMD has dropped all hints at wanting performance lead silicon. Why? There are no margins on the top of a desktop market. Your king of the hill silicon has a premium price and low sales volumes. AMD's graphics division has already shown this strategy in the last 2-3 models.

Also, Dirk was not let go from AMD for anything related to BD. He was let go because he did not get AMD into the mobile chip market and did not want to even at the insistence of the AMD board. That is a fact that anyone can Google. It is well documented on literally every online publication pertaining to business, chip making, and related fields. Personally I think Dirk was spot on in not getting AMD into the mobile market. The margins are complete crap and AMD really does not have the resources to tackle another sector. The AMD board is going to eat it on this one, but I digress.
 
Last edited:
This is just a wild hunch but I'm pretty sure I recognize that use of "I digress" from somewhere.... Charlie?? Is that you?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top