[Techspot] Then and Now: A decade of Intel CPUs compared, from Conroe to Haswell

[AtenRa]

Hmm... 10-15% slower in general (for the same of argument we'll accept that figure, even though most major sites put it much higher), equal to or slightly faster in multi-threaded code, noticeably higher power consumption... why does that sound familiar?

Oh, right! That was how the Prescott Pentium 4 stacked up against the Athlon 64. So, were 2004-era AMD fans wrong to brag about how good the chip was? After all, Prescott had its problems, but it wasn't slower than the Athlon 64 across the board, and it was still fast enough for most people, right?

Yeap, and P4 was a dream to OVERCLOCK reaching more than 4GHz when Athlon was topping up 2.7GHz ??

I still have my 2.4GHz P4 that was able to run at 3.7GHz on air.

 

[Abwx]

Sources have been provided to you in the past. You don't agree and as a result you resort to your usual ad hominem attacks.

But thanks for proving me right

Beyond your own thoughts, that is, right stuck in blank statements and in your usual urban legend...

The only thing MSI confirmed 100% was that particular motherboard VRM was not able to keep up with 125W TDP Bulldozer.

ps: Just because MSI said that motherboard was 125W TDP ready doesnt mean the VRM design was able to cope with 125W TDP CPUs.

A 125W AM3 MB has not the same current requirements as a 125W AM3+ MB wich can supply equal powers but at higher currents..

Of course it s not easy explaining that 125W at 1.4V require 89.3A but that if the CPU use a smaller node with lower voltage, let say 1.25V the MB although supplying 125W will do so with a 100A current, this imply a different design of the CPU power supply.

I guess that in the Intel side there s no such problem, they obsolete their previous MBs...

But i digress, what was it about already.?.

 

[escrow4]

Yeah, this started as a troll thread.

It started as a discussion of an article that devolved. Blame the slavering AMD defenders if anything. Intel is up to Skylake and AMD is still stuck in 2012 with CPUs and 2009 with its chipsets.

 

[AtenRa]

It started as a discussion of an article that devolved. Blame the slavering AMD defenders if anything. Intel is up to Skylake and AMD is still stuck in 2012 with CPUs and 2009 with its chipsets.

Actually in desktop AMD is at 2014.

 

[Phynaz]

The only thing MSI confirmed 100% was that particular motherboard VRM was not able to keep up with 125W TDP Bulldozer.

ps: Just because MSI said that motherboard was 125W TDP ready doesnt mean the VRM design was able to cope with 125W TDP CPUs.

Just because AMD said the 8350 draws 125W really doesn't mean it draws that much. See what I did there?

I'm 100% sure MSI's engineers know about 1,000 times more about motherboard power delivery than you do. I'm also 100% sure they have much greater access to AMD engineering specifications than you do. And they specifically stated that Bulldozer consumed 140W.

Up to you to show MSI was wrong. Proof, not conjecture, which is all your post is.

 

[AtenRa]

Just because AMD said the 8350 draws 125W really doesn't mean it draws that much. See what I did there?

What you did was to confuse TDP with power draw.

And they specifically stated that Bulldozer consumed 140W.

Then they shouldn't allow the Motherboard in question to OFFICIALLY support the FX8350 if that was true.

 

[myocardia]

Now that s what we can call "to beat something", that is 100-200% better perf/$....

You mean utterly decimated, like this?

 

[Phynaz]

What you did was to confuse TDP with power draw.

I knew you'd go there. Deflect all you want, it doesn't change the fact that AMD lied about Bulldozers power requirements.

 

[AtenRa]

I knew you'd go there. Deflect all you want, it doesn't change the fact that AMD lied about Bulldozers power requirements.

You havent provided a single scientific evidence that AMD lied about the power requirements of the Bulldozer. And no, measuring total system power consumption from the wall at idle and full load is not valid to single out the power consumption of the CPU ALONE.

 

[AtenRa]

You mean utterly decimated, like this?

This is like comparing the Pentium G3258 to the FX9590 in MT loads

 

[Phynaz]

So now MSI didn't say Bulldozer draws 140W??

I'm confused.

How about you clear the air by posting facts about what MSI said?

 

[toyota]

NVIDIA PhysX run faster on the Core i7 4770K than on GTX980Ti in Metro Redux ??? i didnt know that, got any review link ??

I would also like to see this.

physx on 980 Ti

Options: Resolution: 1920 x 1080; Quality: Very High; SSAA: On; Texture filtering: AF 16X; Motion Blur: Normal; Tesselation: Very High; VSync: Off; Advanced PhysX: On;
Run 0 (Scene 1 )
Total Frames: 14102, Total Time: 171.2015 sec
Average Framerate: 82.41
Max. Framerate: 229.46 (Frame: 11559)
Min. Framerate: 19.40 (Frame: 9026)


physx on the 4770k

Options: Resolution: 1920 x 1080; Quality: Very High; SSAA: On; Texture filtering: AF 16X; Motion Blur: Normal; Tesselation: Very High; VSync: Off; Advanced PhysX: On;
Run 0 (Scene 1 )
Total Frames: 14749, Total Time: 171.2853 sec
Average Framerate: 86.14
Max. Framerate: 194.06 (Frame: 11995)
Min. Framerate: 10.61 (Frame: 57)


and before anyone says anything silly about the min framerate, it is irrelevant in this fly by bench as it measures in fractions of second and is not very consistent nor does it apply to the actual game.

 

[AtenRa]

So now MSI didn't say Bulldozer draws 140W??

I'm confused.

I can say the moon is red and bleeding, where is the evidence ???

Just because a single MSI motherboard model had a problem throttling the FX8350 in order to keep the VRMs in specs (admitted it was done by MSI themselves) doesnt mean the CPU is drawing more than it is rated.

MSI haven't provided any scientific evidence to demonstrate that the CPU was drawing more than it was rated. They just said the CPU draws 140W and immediately you and certain other members took if at face value.
If it was an Intel CPU im sure the Motherboard maker would be at fault no questions asked. :whiste:

 

[ShintaiDK]

Did AMD ever release specs for the FX series? Or is it just 125W at unknown voltage at unknown temperature?

http://products.amd.com/en-gb/DesktopCPUDetail.aspx?id=809

 

[AtenRa]

physx on 980 Ti

Options: Resolution: 1920 x 1080; Quality: Very High; SSAA: On; Texture filtering: AF 16X; Motion Blur: Normal; Tesselation: Very High; VSync: Off; Advanced PhysX: On;
Run 0 (Scene 1 )
Total Frames: 14102, Total Time: 171.2015 sec
Average Framerate: 82.41
Max. Framerate: 229.46 (Frame: 11559)
Min. Framerate: 19.40 (Frame: 9026)


physx on the 4770k

Options: Resolution: 1920 x 1080; Quality: Very High; SSAA: On; Texture filtering: AF 16X; Motion Blur: Normal; Tesselation: Very High; VSync: Off; Advanced PhysX: On;
Run 0 (Scene 1 )
Total Frames: 14749, Total Time: 171.2853 sec
Average Framerate: 86.14
Max. Framerate: 194.06 (Frame: 11995)
Min. Framerate: 10.61 (Frame: 57)


and before anyone says anything silly about the min framerate, it is irrelevant in this fly by bench as it measures in fractions of second and is not consistent.

Nice thanks,

The Core i7 4770K is at 4.4GHz as you have it in your sing ??

So, they run PhysX in CPU in the review in order to have such a big impact on the Core 2 Quad. If they would use GPU PhysX there would be a smaller impact and the game with the Core 2 Quads would be perfectly playable. ??

 

[Abwx]

Did AMD ever release specs for the FX series? Or is it just 125W at unknown voltage at unknown temperature?

http://products.amd.com/en-gb/DesktopCPUDetail.aspx?id=809

They release much better than specs on paper datasheets, they provide the Spice models of their products, just put the model in the simulator, connect a parametrised power supply, set the clock frequencies with a generator and run all the simulations you wants, dynamic behaviour under load is perfectly modelised within 1%...

I guess that people should do some howmework before posting non sense, currently without such models it is is impossible to reliably design whatever is using a few dozen electronic components, let alone a MB, this has become so strategical to get products rapidly on market that even basic models of generic components like VRM Mosfets are often crypted and no more provided to whom is not a customer...

 

[Phynaz]

Just because a single MSI motherboard model had a problem throttling the FX8350 in order to keep the VRMs in specs (admitted it was done by MSI themselves) doesnt mean the CPU is drawing more than it is rated.

That's exactly what it means.

AMD : design for 125W of power consumption.
MSI: they actually draw 140W of power.

The above are all facts.

 

[Abwx]

The Core i7 4770K is at 4.4GHz as you have it in your sing ??

From 4GHz to 4.4GHz a Haswell dissipated power typicaly increase by 40-50%, and the chip is not adapted to dissipate huge amounts of power given the die reduced area, this mandate paradoxaly even better cooling than for a FX, yet another source of savings, isnt it...

 

[AtenRa]

That's exactly what it means.

AMD : design for 125W of power consumption.
MSI: they actually draw 140W of power.

The above are all facts.

Whatever float your boat.
You dont seem to understand that other MSI Motherboard models didnt have that problem. It was only with a specific cheap model that MSI shouldn't officially support the CPU in the first place.

 

[Abwx]

That's exactly what it means.

AMD : design for 125W of power consumption.
MSI: they actually draw 140W of power.

The above are all facts.

That s all words, a link if possible otherwise it s rumours that eventualy started at MSI but are contradicted by any review of an actual retail sample.

Hardware.fr got 125W under prime 95, actualy the only CPUs that were above their rated TDP were the Haswell 4770K/4790K, 30% above to be precise if 256K FFT are used in Prime 95.

You want the link or a translation, or both.???.

That s fact that are verifiable contrary to your urban legends..

 

[Phynaz]

Weather or not MSI chose to support the CPU is irrelevant. That doesn't mean AMD didn't lie about power consumption.

There are many reviews that show AMD lied.

Wasn't this around the same time that AMD started talking about average power instead of TDP because they were afraid to tell the truth?

 

[Abwx]

Weather or not MSI chose to support the CPU is irrelevant. That doesn't mean AMD didn't lie about power consumption.

There are many reviews that show AMD lied.

Wasn't this around the same time that AMD started talking about average power instead of TDP because they were afraid to tell the truth?

And Atenra own an FX and has checked, what is your urban legends based assumptions worth in comparison..?.

You realize that you are pulling and pushing hot air at threads lengths..?.

You couldnt even find a single review that support your myths, even AT review is suspicious apparently, are you not tired of your extreme bias..?.

 

[SlowSpyder]

You talk about overclock, but you forget to overclock one part. in that case I can make a G3258 looks REALLY good vs some stock AMD CPUs.

Also you completely forget TCO. Not to mention the 4.4Ghz OCed 150-200W 8320E spaceheater still cant compete in the 95%+ cases.

For what it's worth, I can under-volt @ 4.4GHz and use less than 100 watts in Cinemark with all eight threads pegged.

 

[SlowSpyder]

You mean utterly decimated, like this?

Even in these worst case examples, the FX 8350 still provides more performance per dollar than a 4770k or 4790k. Which obviously is the problem, AMD would much rather have been able to get $300+ for these, there is a reason they are cheaper... AMD's design is just too unbalanced compared to Intel. But, in my opinion that doesn't necessarily make it a bad CPU.

 

[mrmt]

Wasn't this around the same time that AMD started talking about average power instead of TDP because they were afraid to tell the truth?

Nope, ACP was created circa 2007 when Opteron got spanked by Clovertown Xeon, so they tried to bring this smoke screen to the market. With Bulldozer the deception reached new heights, as they didn't even bother to release the thermal datasheet, as they used to do since the K8 era.

Btw, don't expect resellers to ever admit the products they sell operate outside the rated specifications, as they may be liable under consumer laws to refunds, damages and penalties, along with the manufacturer.