AMD back in gear, Centurion FX

[CHADBOGA]

IF I had to choose a rig and based on AMD's superiority in highly threaded tests, I would go for AMD as this would best work as an encoding rig I suppose.

Even if it was behind in low threaded tasks, I would venture and say the performance cannot be that bad that the rig is unusable in low threaded tasks.

edit: Of course, if I have the bucks and can just choose 'the best', I'd definitely go Intel. On a budget, I'd say AMD.

Right now and for who knows how many years, most people will run into trouble with their Quad Core CPU's when their single threaded performance just isn't good enough.

I can always wait a few more seconds or minutes for a video to encode, but if my single threaded performance is lacking, that is when games will start to be frustrating experiences.

 

[guskline]

Kudos bgt, nice test results. What instrument did you use for measuring wattage?

 

[CHADBOGA]

I just dont hope the forum suffers the same fate due to that.

What a cautionary tale XS provides to forums everywhere, of what happens when you pander to fanatics, to show your "fairness".

Their dead forum is exactly what they deserve for their idiocy.

 

[Abwx]

Funny to see the die hard intel fans trashing this thread
and derailing it while talking of another site wich is/was
supposedly full of fanatics...

 

[CHADBOGA]

Funny to see the die hard intel fans trashing this thread
and derailing it while talking of another site wich is/was
supposedly full of fanatics...

Being highly skeptical of the viability of a 5Ghz CPU from AMD at this time, is not trashing.

What people like you want, is for everyone to disengage their thinking and give glorious praise to AMD for a product that likely will never emerge this year, or if it does, will sell poorly for all the reasons already explained in this thread.

That kind of cult like behaviour is just plain silly.

 

[Charles Kozierok]

Funny to see the die hard intel fans trashing this thread
and derailing it while talking of another site wich is/was
supposedly full of fanatics...

I'm not an Intel fan. I'm a reality fan.

Reality right now is that Intel is leaving AMD in the dust. I'm actually not happy about that, but it is what it is, and I'm not going to bend over backwards to make Intel look worse and AMD look better just to fool myself into thinking there's a competition going on here when there really is not.

 

[Abwx]

If this CPU is actualy released it will forcibly be compatible
with at least the best AM3+ MBs , none of wich is specified
for more than 196W continous.

Foremost , since it is said to work on air it cant be higher
than those 196W if use is to be more than a superpi bench.

 

[Idontcare]

Thanks for posting those test numbers.

Tested power usage of FX8350:
..
8/245W/174W/8103
..
Tested power usage of i7-3770K:
8/128W/93W/9424

Er.. wow.

You aren't supposed to point out that it is basically a 100W delta...folks don't like that. You are supposed to blame the mobo, the ram, the HDD, the PSU, the kill-a-watt, the moon, the sea, the mouse in the wall...anything but the CPU.

Mind you the power-company is not fooled, they will charge you every penny of that extra power usage, they live in reality I guess. Must be nice.

 

[guskline]

You aren't supposed to point out that it is basically a 100W delta...folks don't like that. You are supposed to blame the mobo, the ram, the HDD, the PSU, the kill-a-watt, the moon, the sea, the mouse in the wall...anything but the CPU.

Mind you the power-company is not fooled, they will charge you every penny of that extra power usage, they live in reality I guess. Must be nice.

Agreed. bgt's numbers not only point out the 100 w gap between the 8350 and the 3770k but show that the 3770k running 100watts cooler performs significantly better.

 

[Idontcare]

If this CPU is actualy released it will forcibly be compatible
with at least the best AM3+ MBs , none of wich is specified
for more than 196W continous.

Foremost , since it is said to work on air it cant be higher
than those 196W if use is to be more than a superpi bench.

Well now you are singing a different tune. It is a very different thing to make the argument that AMD will bin out special golden samples that can operate with such a low Vcore and have such a low static leakage and operating temperature as to manage to fit inside a 200W TDP at 5GHz.

Anything is possible if they want to search for those elusive chips.

I dare say it is a far better use of their take-or-pay monies as well. At least make GloFo produce all those wafers, since you are paying for them anyways, and then bin out the one golden chip to be found every 10 wafers or so and sell them for a princely sum as a means to recoup some small percentage of the take-or-pay fees that you are going to pay anyways.

It can make financial sense, given the specific corner AMD finds itself backed into.

 

[Abwx]

Well now you are singing a different tune. It is a very different thing to make the argument that AMD will bin out special golden samples that can operate with such a low Vcore and have such a low static leakage and operating temperature as to manage to fit inside a 200W TDP at 5GHz.

Not at all , i didnt say that it is 196W but that it will fit in this
enveloppe , probably in the 140-150W range.

You are basing your opinion in your chip experiment that you hold as being
representative while it is obvious that there s chips that have way
better specs as the one i quoted wich do 4.7G at 1.47V and 310W/Wprime/wall , far from your 1.56V for 4.8G.

The link provided a page earlier show even a 5.1G chip stable
with 1.398V so it s quite possible that AMD can bin a few hundreds ,
or let say one thousand out of 500 000/quarter (that would
make 0.002% of a quarter prod.) to pick the good ones.

 

[guskline]

Funny thing about this thread is that IF the AMD Centurion ever sees the light of day, and I think that will be a BIG IF, Intel will have already released the Haswell 4770k.

I doubt even with big demand it will exceed $400 in price- Half of the Centurion price.

Want to wager which cpu will win the performance crown?

Haswell I7-4770k vs AMD FX8??? Centurion?

My money is on Intel.

 

[purefun1965]

What a cautionary tale XS provides to forums everywhere, of what happens when you pander to fanatics, to show your "fairness".

Their dead forum is exactly what they deserve for their idiocy.

been a member over at xs since 2004 and this is quote is bs. infact i sent charles aka fugger a heads up.

 

[Gikaseixas]

just wanted to say this:

Intel rules the land right now, no there's no contest but to be fair to the AMD guys, i can push my FX to 5.0Ghz with 1.48 volts. I just don't use it like that because there's no gain in performance from 4.7 to 5.0 on these chips, at least IMO.

Last, if they release this i believe it will be a newer revision. There's no way they can release this thing without burning mb's at a first oc attempt.

 

[CHADBOGA]

been a member over at xs since 2004 and this is quote is bs. infact i sent charles aka fugger a heads up.

No it is the truth.

Anyone who was around there in 2008 will know it was once a vibrant forum.

 

[Idontcare]

Not at all , i didnt say that it is 196W but that it will fit in this
enveloppe , probably in the 140-150W range.

You are basing your opinion in your chip experiment that you hold as being
representative while it is obvious that there s chips that have way
better specs as the one i quoted wich do 4.7G at 1.47V and 310W/Wprime/wall , far from your 1.56V for 4.8G.

My voltages were physically measured, not relying on CPUz reports, and I used a prior version of Prime95 that actually generates more power-consumption and heat than the more recent version.

Its all documented in my FX8350 thread. If you aren't aware of the situation with prime95 and piledriver then you should read the threads in the mersenne forums in which the authors of prime95 discuss the fact that piledriver has such horrible latency with AVX and SSE2 instructions that although they (prime95) can use them the performance is actually lower (and the heat is as well). So you go backwards, regress, if you try and force software to use the latest instructions in the AMD ISA.

I can make my power numbers look great too, just find software that stresses the CPU less and use a mobo that poorly measures/reports the Vcore.

Surely you aren't so easily fooled by anecdotal reports on the internets, right? I give you more credit than that.

(put it a different way, do you really think MSI engineers went to time, trouble, and expense to throttle the FX-8350 if it really doesn't have power-consumption issues?)

The link provided a page earlier show even a 5.1G chip stable
with 1.398V so it s quite possible that AMD can bin a few hundreds ,
or let say one thousand out of 500 000/quarter (that would
make 0.002% of a quarter prod.) to pick the good ones.

"Stable" means very different things to different people.

None of our enthusiast stress testers actually test the functional/operational stability of every instruction in the ISA under load-conditions.

There is a reason, a good reason, why AMD bins out the various CPUs to request the voltages they request of the BIOS and VRMs.

Its not like AMD wants its chips to use 125W and 1.4V. If they could bin 5GHz chips which only require 1.4V then surely we'd all be reading reviews of an FX-8370 that runs at 4.5GHz (5GHz turbo) months ago.

But finding one guy on the internet who undervolts his 8350 to the point of being questionably stable while running one stress tester is not exactly "evidence".

Don't you remember those early reports of crazy low Vcores with Sandy Bridge chips at 5GHz in these forums? (was it member Smobkalit?) Only we (and he) found that once he actually used the right version of the stress tester then his system crashed unless he raised the voltage back up to something normal and expected?

One-off reports are one-off for a reason, extraordinary claims require extraordinary proof. I see no proof, just see the claims. From him, and you.

 

[Puppies04]

I dare say it is a far better use of their take-or-pay monies as well. At least make GloFo produce all those wafers, since you are paying for them anyways, and then bin out the one golden chip to be found every 10 wafers or so and sell them for a princely sum as a means to recoup some small percentage of the take-or-pay fees that you are going to pay anyways.

I never thought about it that way. I always hear that the CPUs that come from the centre of the die tend to be better than the ones from the edge, taking this into consideration AMD wouldn't even need to bin half the CPUs.

 

[guskline]

just wanted to say this:

Intel rules the land right now, no there's no contest but to be fair to the AMD guys, i can push my FX to 5.0Ghz with 1.48 volts. I just don't use it like that because there's no gain in performance from 4.7 to 5.0 on these chips, at least IMO.

Last, if they release this i believe it will be a newer revision. There's no way they can release this thing without burning mb's at a first oc attempt.

Excellent post! I run my 8350 at 4.6Ghz for the same reason.

 

[insertcarehere]

probably in the 140-150W range

So you are saying that, somehow, AMD will be able to get a not insignificant amount of 4 module vishera @5ghz operating at "140-150W" range, don't be surprised when people dispute this point heavily, considering how every single review on piledriver as shown otherwise.

 

[purefun1965]

No it is the truth.

Anyone who was around there in 2008 will know it was once a vibrant forum.

worlds best overclockers post there. I dont see them post here. I wonder why:hmm:

 

[Idontcare]

I always hear that the CPUs that come from the centre of the die tend to be better than the ones from the edge, taking this into consideration AMD wouldn't even need to bin half the CPUs.

That's a myth that isn't born out in reality. Trust me on that.

Center-to-edge yield variations are a real thing, having center die better than edge die (so-called bulls-eye pattern) can happen but is not required to happen (is not a foregone conclusion).

As the process is tuned, and they always are, the bulls-eye pattern can become a donut pattern (poorer dies are at the edge and the center), or tuned such that the edge yields the best dies.

No one, and I do mean no one, intentionally runs their process nodes such that the cherry chips come from the center of the wafer as that is the absolute least optimal way to produce your products (since the areal-situation so strongly favors tuning for edge die yields).

These are the sorts of myths that I have recognized, and accepted, as simply never dying out. They are so catchy, and so easily remember, that no amount of snopes-like work on my part will ever vanquish the popularity of the myth. (very much like process node labels versus actual dimensions )

But at least you will now know something closer to the truth, should you take my word on it as written here. (and I recognize that you have little reason to take my word on it either, you have no proof that I am who I claim to be, so skepticism on your part is reasonable, if not expected, as well)

 

[Idontcare]

worlds best overclockers post there. I dont see them post here. I wonder why:hmm:

I felt bad when AMD decided to turn them into shills for its "over 8GHz world record on bulldozer!!!!" fest. Lot of guys lost credibility, and touch with their fanbase, when they participated in that marketing stunt while keeping mum on the disastrous performance issues that the clockspeed could not overcome.

Glad to have world's best OCers existing in the world, but I don't trust a one of them to actually generate information for me that I can use to do anything with. They all have their masters to pay heed to, and do they ever.

 

[Chiropteran]

I own 2 FX-8120 machines and now an i7-3770. In regular usage I generally can't tell the difference. One FX-8120 based system has an SSD and feels noticeably faster than the i7, actually, but for the most part they all feel about the same. I've measured power usage and seen the FX system use 10-20 more W in idle and general usage, but that difference equates to an extra $1.30 per month in electricity cost, *if* I run the FX system 24 hours a day. I do not feel feel the need to expend energy and money replacing the FX systems with more efficient intel systems.

If I was building another system today, I'd choose between an FX-8350 bundle at Microcenter ($239.98 for CPU and Motherboard) or $284.98 for the i5-3570k equivalent bundle. I'm not sure which one I would buy, it would depend on the exact usage of the new build, but I know that the greater power consumption of the FX system wouldn't be a concern at all- the $45 price premium of the intel system will more than cover the power consumption difference for the next 3 years or more. In fact for a light usage person who turns the PC off after using it, if you figure 4 hours a day of usage, it will take 18 YEARS for the i5 to reach price parity.

The point is, the difference in power consumption is absurdly overrated. It really doesn't make a big deal in reality, it's just a popular topic to argue about from the intel fans because they happen to have an advantage right now.

 

[AtenRa]

(put it a different way, do you really think MSI engineers went to time, trouble, and expense to throttle the FX-8350 if it really doesn't have power-consumption issues?)

I believe that we can also ask if the MSI Motherboard's VRM in question can actually handle 125W TDP CPUs.

From Anandtech FX8350 review, we have a maximum Power consumption (x264) of the wall for the entire system at 195.2W, now take off 19,6W for the North Bridge(990FX). Idle power measured to be 74.2W so we have,

195.2-19.6-74.2 = 101,4W

Edit: Anandtech used a motherboard with 990FX not 970

If we take off the PSU and VRM efficiency then we arrive at the conclusion that the CPU alone uses less than 100W when running a real everyday application stressing all 8 threads like x264.

ps: just to remind everyone that TDP is not the maximum power consumption of the AMD/Intel CPUs.

The thermal design power (TDP), sometimes called thermal design point, refers to the maximum amount of power the cooling system in a computer is required to dissipate. The TDP is typically not the most power the chip could ever draw, such as by a power virus, but rather the maximum power that it would draw when running "real applications"

 

[Enigmoid]

I believe that we can also ask if the MSI Motherboard's VRM in question can actually handle 125W TDP CPUs.

From Anandtech FX8350 review, we have a maximum Power consumption (x264) of the wall for the entire system at 195.2W, now take off 19,6W for the North Bridge(990FX). Idle power measured to be 74.2W so we have,

195.2-19.6-74.2 = 101,4W

Edit: Anandtech used a motherboard with 990FX not 970

If we take off the PSU and VRM efficiency then we arrive at the conclusion that the CPU alone uses less than 100W when running a real everyday application stressing all 8 threads like x264.

ps: just to remind everyone that TDP is not the maximum power consumption of the AMD/Intel CPUs.

True

But

120-60=60 watts for the i7-3770k

101-58=43 watts for the i5-3570k.

Z77 chipset uses about 7 watts.

That cpu is not using 0 power at idle. The difference between idle and load is 100 watts, not the chip using at load 100 watts.