• We should now be fully online following an overnight outage. Apologies for any inconvenience, we do not expect there to be any further issues.

Phoronix : FX 8350 competitive with Core i7 3770K under Linux.

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Vic Vega

Diamond Member
Sep 24, 2010
4,535
4
0
ah NM.. still 4.2ghz vs 3.9ghz still looks 25% slower.

I'm trying to decide how this even matters. Are you suggesting if I buy a CPU I should clock it up or down to match the closest competing model from the competitor to make it "equal"?
 

9enesis

Member
Oct 14, 2012
77
0
0
it would be interesting to see FX8150 @ 4Ghz vs FX8350 @ 4Ghz to sort out the IPC gain(if any).
 

sefsefsefsef

Senior member
Jun 21, 2007
218
1
71
The overclocked Vishera numbers are meant to be compared to the stock Vishera numbers. It's not a conspiracy.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
it would be interesting to see FX8150 @ 4Ghz vs FX8350 @ 4Ghz to sort out the IPC gain(if any).
I'm sure that has been run. It's just that we have to wait for the nda to expire.
 
Last edited:

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
The overclocked Vishera numbers are meant to be compared to the stock Vishera numbers. It's not a conspiracy.

It's not a conspiracy, but it's laziness.

For things that the 8350 is well suited for, it would be useful to see it directly compared OC vs. OC against 3570K and 3770K.
 

ctsoth

Member
Feb 6, 2011
148
0
0
Maybe during all this restructuring AMD bought up rabbits, some fine top hats, and hired a score of magicians.
 
Aug 11, 2008
10,451
642
126
To be fair, it was the conclusion that the author of the article put forth. Quibbling with it seems reasonable based on the data, but it had its source in the article.

I mean, Anandtech.com never has a conclusion we disagree with, right? :)

I did acknowledge that he linked the whole article. However, he could also have quoted the less decisive conclusion where the article said it is "a rather competitive value" IF
"you are commonly engaging in a workload where AMD CPUs do well". Kind of like saying your football team would be undefeated if you only counted the games that you won.
 

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
I did acknowledge that he linked the whole article. However, he could also have quoted the less decisive conclusion where the article said it is "a rather competitive value" IF
"you are commonly engaging in a workload where AMD CPUs do well". Kind of like saying your football team would be undefeated if you only counted the games that you won.

Well, I could see the idea that possibly if 8350 hits around $200ish soon, and can deliver 90% of the 3770K OC performance for heavily-threaded tasks, that it might be a decent option.

However, with this stupid Phoronix review, they have OC 8350 vs. Stock 3770K testing. Idiotic.

Look at the title of this thread : 8350 competitive with 3770K. How do we even know given their testing? Because people buy tons of Intel K-series chips and run them stock, but these same people would overclock an 8350? Ludicrous.
 

Olikan

Platinum Member
Sep 23, 2011
2,023
275
126
As for tomorrow's Vishera launch, I have a feeling you'll be eating your words.

quote from SKYMTL (hardware canucks reviwer) in extremsystems forum...

:thumbsup:
 

USER8000

Golden Member
Jun 23, 2012
1,542
780
136
Last edited:

hokies83

Senior member
Oct 3, 2010
837
2
76
I'm trying to decide how this even matters. Are you suggesting if I buy a CPU I should clock it up or down to match the closest competing model from the competitor to make it "equal"?

At Stock clocks it is 15% Slower..

Now think of all the windows based programs.. gonna be a even bigger Gap..

Looks like another Amd Fail to me. Or more holes in the sinking Amd Ship.
 

Greenlepricon

Senior member
Aug 1, 2012
468
0
0
I can't count the number of places you can find a Core i7 3770k OC benchmark. Yeah more work for you. I see stock 8350 vs. stock 3770k, along with an idea of how the 8350 looks overclocked. Nothing wrong with that.
 

Vic Vega

Diamond Member
Sep 24, 2010
4,535
4
0
I can't count the number of places you can find a Core i7 3770k OC benchmark. Yeah more work for you. I see stock 8350 vs. stock 3770k, along with an idea of how the 8350 looks overclocked. Nothing wrong with that.

Don't you be brining any rational thought into this thread. This thread is about bashing the AMD CPU and the article creator for having the gall to suggest it's not a complete turd.
 

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
I can't count the number of places you can find a Core i7 3770k OC benchmark. Yeah more work for you. I see stock 8350 vs. stock 3770k, along with an idea of how the 8350 looks overclocked. Nothing wrong with that.

There are indeed a ton of 3770K OC benches out there. However, the test suite that Phoronix has laid out for this particular review is unique compared to most given the software used, the revisions, and so on. To not include the 3770K OC results within this context is laziness at best, because a true apples to apples 8350OC vs 3770KOC with identical software/revisions/settings is not accessible. One can only approximate by looking at 3770K testing in different circumstances.

I'm not sure why someone would try to say that's not ideal. After all, the 8350 pricing is very attractive for heavily multithreaded situations. A 3770K is about $100 more, so if the 8350OC is still competitive in many situations compared to a 3770OC, WHY NOT INCLUDE THAT IN THE TESTING.

LAZY

LAZY

LAZY
 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
LAZY

LAZY

LAZY

Lazy who didnt pay attention to all the post in this thread....

not sure if true...but good
http://www.cowcotland.com/articles/1199/test-processeur-amd-fx-8350.html

jeux.jpg
 

Greenlepricon

Senior member
Aug 1, 2012
468
0
0
Don't you be brining any rational thought into this thread. This thread is about bashing the AMD CPU and the article creator for having the gall to suggest it's not a complete turd.

Sorry forgot I'm supposed to start raging in here against all AMD!

Anyway still waiting for the benchmarks on windows. Hopefully it's a big enough improvement to kick my bulldozer out of my rig. Too bad it doesn't look like it will have a chance against Intel since they're still behind by too much for now. The only competitive point I'm seeing right now are for those special workloads and its price tag, which will hopefully be looking nice for this holiday.
 

Schmide

Diamond Member
Mar 7, 2002
5,745
1,036
126
Just glancing at the numbers, it looks like efficiency went down by 3.5%?

Edit: Ok further looking at the numbers the extreme turbo 600mhz of the fx-8150 vs 200mhx of the fx-8350 seems to be skewing my results.

Taking the x264 benchmark, since it uses all cores, OC efficiency seems to be 80% and adjusted IPC efficiency improvement seems to be 6.5%.

Edit2: Other encoding looks.

VP8 Seems to have OC efficiency of 93.5% and IPC efficiency improvement of 1.45%
Fmpeg 55% -0.18%
Opus 68% -2.06%
 
Last edited:

Olikan

Platinum Member
Sep 23, 2011
2,023
275
126
1 year of waiting and 8 cores to match intel's dual core who has 1/4 the power consumption and costs half. Still fail at the games section, improved than before though ():)

if bulldozer thought us something...
Probably the 2M\4core won't see much performance hit, but power comsumption will drop nicely...
 

sequoia464

Senior member
Feb 12, 2003
870
0
71
Well, I could see the idea that possibly if 8350 hits around $200ish soon, and can deliver 90% of the 3770K OC performance for heavily-threaded tasks, that it might be a decent option.

Jury is still out on performance - i bet that will be debated heavily for quite a while - but this slide seems so confirm the 8350 starting price of $195.

http://forums.anandtech.com/showthread.php?t=2278251

Lots of posts here today, in case you missed it.
I chose poorly a year ago, hoping this is a viable route for me.
 

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
There are indeed a ton of 3770K OC benches out there. However, the test suite that Phoronix has laid out for this particular review is unique compared to most given the software used, the revisions, and so on. To not include the 3770K OC results within this context is laziness at best, because a true apples to apples 8350OC vs 3770KOC with identical software/revisions/settings is not accessible. One can only approximate by looking at 3770K testing in different circumstances.

I'm not sure why someone would try to say that's not ideal. After all, the 8350 pricing is very attractive for heavily multithreaded situations. A 3770K is about $100 more, so if the 8350OC is still competitive in many situations compared to a 3770OC, WHY NOT INCLUDE THAT IN THE TESTING.

LAZY

LAZY

LAZY

They may not have access to the 3770K at the time.