(gamersnexus) AMD: "FX is Not EOL" & Why What We Need in a CPU is Changing

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

NaroonGTX

Member
Nov 6, 2013
106
0
76
The only reason people kept screaming that it was fake is because they didn't wanna believe what they saw. "Oh noez!!1! no moar EFF ECKS??!! No DDR4 4 Kareezoh>??!! /wrists"
 

mikk

Diamond Member
May 15, 2012
4,304
2,391
136
This is often the reason yes. Every leak get its fake screaming.
 

Leyawiin

Diamond Member
Nov 11, 2008
3,204
52
91
The only reason people kept screaming that it was fake is because they didn't wanna believe what they saw. "Oh noez!!1! no moar EFF ECKS??!! No DDR4 4 Kareezoh>??!! /wrists"

They probably keep saying it (along with the AMD exec) because there has been nothing officially released by AMD covering that time period. "Leaks" aren't official. Speculation by armchair experts in a hardware forum aren't official. Makes one wonder about the motivation of those who say it isn't fake. Probably at least as suspect for whatever reason.
 

NaroonGTX

Member
Nov 6, 2013
106
0
76
Well I use simple logic to refute its "fakeness". What current info do we have that contradicts that slide?

*waits*

Yep. We know there won't be anything on AM3+ in 2014 since there are no new big-die Opterons coming besides "Warsaw", which is just basically Piledriver with RCM enabled. So if anything, AM3+ would see "one more chip" in the form of a Vishera refresh.

So that leaves 2015, where even still AMD hinted before with its 2013 Server strategy slides that there wouldn't be a XV related chip released either. We don't know this for sure of course, but for now the most logical outcome is that AM3+ is EOL in terms of getting actual new chips.

Other than that, the rest of the info is spot-on. We've had previous leaks mention Carrizo's TDP figures (45W/65W max) and the fact that it would be on FM2+ socket. The map doesn't contradict that at all.

First the slide was uploaded in a thread over at OCN, where of course people said it was fake only because they'd never seen it before (such great logic). That version had a watermark on it since the guy who posted it works for some other tech site. Then recently Sweclockers gets a hold of the exact same sets of slides, this time with their own watermarks or none at all, and the info in the other slides is accurate as well?

I seriously doubt there is some underground secret society dedicated to making one fake slide just to troll everyone. It says 'AMD Confidential' at the bottom which means it was never meant to be shown/released to the public originally. I've also had several sources tell me that they'd seen those same slides at APU '13 in person and another event before that.

So no, I won't take the word of some PR guy who has been wrong in the past as an 'official' confirmation that they were fake. People just need to get over it.
 

jpiniero

Lifer
Oct 1, 2010
16,827
7,271
136
The question really should be whether AMD will do a big-core x86 processor past Vishera.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
146
106
The question really should be whether AMD will do a big-core x86 processor past Vishera.

It doesnt look so. Almost no SR parts either. If they even release EX, it will be FM2+ only and it will most likely be the end of the line for big cores from AMD.
 

NaroonGTX

Member
Nov 6, 2013
106
0
76
You're being needlessly pedantic.

And you're being needlessly petty. I'm merely going by AMD's own definition. "Big core" doesn't necessarily mean "FX/Opteron". They're the same exact x86 cores used across the higher-end APU's, the FX, and the Opteron (of which the latter two share the same die anyway.)

How hard is it to simply say FX or "big-die" instead?
 

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136
NaroonGTXis correct in saying Carizzo has a big core in it since Excavator IS a big core family. It might be the last AMD's "big" core but that's what it is.
 

NaroonGTX

Member
Nov 6, 2013
106
0
76
LOL what blunder? You're the only one getting upset over it. All I did was offer a correction, the fact BD, PD, SR, and EX are all "big cores" and they are in fact inside the current and upcoming APU's. Whether or not AMD will continue with big cores beyond Carrizo/Basilisk is unknown.
 

Homeles

Platinum Member
Dec 9, 2011
2,580
0
0
LOL what blunder? You're the only one getting upset over it. All I did was offer a correction, the fact BD, PD, SR, and EX are all "big cores" and they are in fact inside the current and upcoming APU's. Whether or not AMD will continue with big cores beyond Carrizo/Basilisk is unknown.
If you can't see how your embarrassing failure to comprehend jpneiro's post was a blunder, then I suppose I'm wasting my time. I'm not going to derail this thread further.
 

bononos

Diamond Member
Aug 21, 2011
3,938
190
106
I know the extreme edition Intel processors are a different price bracket than mainstream Z87 .

What I saying is that I dont think all AMD FX processors require 220W boards.
AMD FX processors which run at 4.6+Ghz require high end motherboards.
OCed 8320, 8350 & 9000 series.
Heck there are people on Overclock.net running 4.8GHz on the 99FX Pro R2.

Going 4.6+ Ghz on FX series demands an atleast decent quality board.

Plus the most expensive AMD board Crosshair V Formula is 220$.
One of the most popular Haswell mainstream board the Asus Z87 Pro costs 185$ right now.

35$ more is not that bad considering AMD doesn't change sockets every year.
Although AM3+ is EOL right now.

You get 4.8 ~ 5Ghz on Asus Sabertooth board and it costs the same as Z87 Pro and dont want to shell money for halo boards.
It is a myth that you require a 200$ board for FX.

To summarize, as I told you before that mainstream Intel processors when OCed could draw close to 200W and most regular Intel MOBOs handle that. If the manufacturers can use these kinds of components on Intel MOBOs than they can do it for AMD aswell.

Don't you remember the testing that IDC did with the 8350 which showed that it sucked down close to 200W at stock speeds? Or the reviews elsewhere which showed FX cpus using more power than expected.
And what about MSI (iifc) mbs throttling AMD cpus once they exceeded 140W or somewhere there.
Theres reasonable suspicion that AMD might not be following its own spec and also because it has changed its definition of TDP and has also not fully released that sort of info for the FX line, its likely that AMD is quietly releasing hot ones without regard for longevity since its bottom line is hurting so much.
 

NaroonGTX

Member
Nov 6, 2013
106
0
76
If you can't see how your embarrassing failure to comprehend jpneiro's post was a blunder, then I suppose I'm wasting my time. I'm not going to derail this thread further.

I didn't 'fail' to comprehend what he originally meant at all. I knew what he meant when I made my post -- I was merely posting a small correction. YOU were the one who got your jimmies rustled for no good reason. Your original post literally wasn't even necessary. You just brought it up for the sake of an argument.
 

USER8000

Golden Member
Jun 23, 2012
1,542
780
136
I didn't 'fail' to comprehend what he originally meant at all. I knew what he meant when I made my post -- I was merely posting a small correction. YOU were the one who got your jimmies rustled for no good reason. Your original post literally wasn't even necessary. You just brought it up for the sake of an argument.

Agreed. Excavator is a big core. He is getting confused between big core and big die.
 
Aug 11, 2008
10,451
642
126
I think it is obvious that even though not technically worded correctly, people are using "big core" to mean more than 4 core chips and a successor to vishera. If they could release a shrink and get the power consumption down, it might be attractive, since Intel is not making much improvement in the high end desktop either, and stubbornly refuses to release a hex core mainstream chip.
 

NTMBK

Lifer
Nov 14, 2011
10,448
5,832
136
Why don't you just accept responsibility for your blunder? It's quite obvious as to what jp meant.

jp used the wrong term, and Naroon corrected him on it. I see no blunder.

This is a technical forum, it's pretty reasonable to expect people to use the correct terminology.
 

NTMBK

Lifer
Nov 14, 2011
10,448
5,832
136
I think it is obvious that even though not technically worded correctly, people are using "big core" to mean more than 4 core chips and a successor to vishera.

In which case people are using it incorrectly.

I could start using "potato" to refer to GPUs, if I felt like it. Doesn't change the fact that I would be incorrect.
 

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136
AMD had presented their core strategy years ago. They clearly said, in no uncertain terms, that big cores were codenamed after big machinery(bulldozer,piledriver,steamroller, excavator) and small cores were codenamed after cats(bobcat,jaguar,puma,xxxxx).

EX core is big core and chips based on it can be called whatever AMD likes. This doesn't change the fact that those chips are based on the big core, namely Excavator. Similar goes for small cores. "Big chip" is not the same thing as big core, never was.
 

Stone Rain

Member
Feb 25, 2013
159
0
0
www.stonerain.us
AMD may not be officially killing the FX lineup; but if the line isn't going to updated until 2015, it might as well be dead, because FX is already a questionable buy compared to Haswell in nearly every usage case. Examples:

Super-budget usage: The FX-4xxx CPU's aren't really a viable buy; the performance per dollar isn't worth buying, save thirty more bucks and move up a notch.

Budget usage: The FX-6300 effectively has one target market: budget gamers. Businesses that don't need GPU horsepower would be silly to buy an FX CPU and a GPU when they could use an AMD APU or any Intel CPU with HD graphics. The problem is, the i3-4130 trashes the 6300 in single-core performance, which means any non-game application like a web browser, etc, is going to run significantly smoother. Also, in games, the performance advantage of the 6300 isn't massive anyway. The other advantage the Haswell i3 has is that the 1150 platform has at least one more generation that will be used on it, coming in 2014. The AM3 platform does not. So in that segment, call it a tie if you must.

Midrange segment: The i5 4670k has 2% lower multithreaded performance than the FX-8350 and 53% faster single threaded performance, plus for business use cases the i5 has an integrated GPU. Also, again we have the 1150 upgrade coming in 2014.

High end segment: The 8350 isn't a competition at all with the i7 or Xeon E3 lineup, and the FX 9590 is a terrible buy because it still lags behind the 4770K, and it will set your house on fire without water cooling.

I'm not trying to be a fanboy here; but am I wrong in saying that AMD with the exception of its' APU's is already a questionable/bad buy and will only become a worse buy after the successor to Haswell crops up?
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
Don't you remember the testing that IDC did with the 8350 which showed that it sucked down close to 200W at stock speeds?

IDC only measured TOTAL SYSTEM power Usage, not CPU usage. It only showed the power difference from Idle to Full Power running Linx.


And what about MSI (iifc) mbs throttling AMD cpus once they exceeded 140W or somewhere there.

Those MSI motherboards had low-end VRM implementations not designed for 125W CPUs.

Theres reasonable suspicion that AMD might not be following its own spec and also because it has changed its definition of TDP and has also not fully released that sort of info for the FX line, its likely that AMD is quietly releasing hot ones without regard for longevity since its bottom line is hurting so much.

AMD hasnt changed its definition of TDP.
 

mrmt

Diamond Member
Aug 18, 2012
3,974
0
76
Theres reasonable suspicion that AMD might not be following its own spec and also because it has changed its definition of TDP

AMD change their TDP specs, they simply didn't disclose the data for the FX series so nobody can verify their 125W claims, which are by all means false.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
The problem is, the i3-4130 trashes the 6300 in single-core performance, which means any non-game application like a web browser, etc, is going to run significantly smoother.

Really ??? :rolleyes:

What about FX6300 OCed to 4.6-5.0GHz completly trashes the Core i3 in every Multithreaded application out there ?? Ahh yes i forgot, we only care about single-thread and Web Browsing :p