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

[AtenRa]

Haha, no. Funny joke tho.

That is total SYSTEM peak power, it has nothing to do with CPU 125W TDP.

 

[MiddleOfTheRoad]

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.

Not really. The only way the G3258 can look good is by completely ignoring the minimum frame rates for pretty much every modern game --a dual core pretty much stutters out on everything now coming out.

BTW, I own a G3258 with a stable overclock to 4.3 Ghz -- it's an amazing chip for playing 2013 and earlier games.... Completely chokes on anything Multi-threaded.
A cheap 760K/860K is a much better investment for modern gaming.... 4 threads is required for today's games (i3 or better for Intel).

 

[Headfoot]

Core 2 Duo/Quad is just not enough

The jump from the top end C2Q to the i5-760 is unexpectedly massive there. On the other hand, it shows how good of a deal the ~$25-40 i5 750 + $50 or less p55 mobo combo is that you can find on ebay or forums used around here. Especially because those i5-750s overclock great. I did exactly this plus a $100 used 7950 Boost and I had a competent guest machine for 1080p.

 

[ShintaiDK]

That is total SYSTEM peak power, it has nothing to do with CPU 125W TDP.

All I see is excuses, not you showing a 4.4Ghz 8320E is 125W or lower as you claimed.

We already know the 8350 stock is 140W.

 

[ShintaiDK]

Not really. The only way the G3258 can look good is by completely ignoring the minimum frame rates for pretty much every modern game --a dual core pretty much stutters out on everything now coming out.

BTW, I own a G3258 with a stable overclock to 4.3 Ghz -- it's an amazing chip for playing 2013 and earlier games.... Completely chokes on anything Multi-threaded.
A cheap 760K/860K is a much better investment for modern gaming.... 4 threads is required for today's games (i3 or better for Intel).

If you want to talk about gaming. Then what do you call a FX, even at 4.4Ghz? Competitive? No. Slowpoke as hell? Yes.

 

[ShintaiDK]

The jump from the top end C2Q to the i5-760 is unexpectedly massive there. On the other hand, it shows how good of a deal the ~$25-40 i5 750 + $50 or less p55 mobo combo is that you can find on ebay or forums used around here. Especially because those i5-750s overclock great. I did exactly this plus a $100 used 7950 Boost and I had a competent guest machine for 1080p.

It has to be said tho, that the IMC looks to be the main reason for the change. not actual processing power. But rather memory access behaviour by the game.

But ye, old Lynnfields still rock.

 

[myocardia]

More benchmarks that are not years old, or rather that use very recent versions :

http://www.hardware.fr/focus/99/amd-fx-8370e-fx-8-coeurs-95-watts-test.html

Wow, so AMD's fastest 8-core can't even beat Intel's fastest quad-core, at anything, and you think that is something of which to be proud? Really? Have you seen how horribly Intel's 6 and 8-core CPUs destroy AMD's fastest CPUs? It's pathetic.

 

[crashtech]

I blame the OP for mentioning AMD at all.

 

[Abwx]

We already know the 8350 stock is 140W.

100% fud...

Who is your "we", btw, the usual suspects that constantly bash AMD..?.

 

[Phynaz]

I blame the OP for mentioning AMD at all.

Yeah, this started as a troll thread.

 

[Phynaz]

That is total SYSTEM peak power, it has nothing to do with CPU 125W TDP.

The rest of the system using over 110 watts is highly unlikely.

 

[Abwx]

Wow, so AMD's fastest 8-core can't even beat Intel's fastest quad-core, at anything,

Do you know how to read a graphs i posted..?.

Here the average of thoses tests :

What is this "anything you re talking about"?.

Your i5 should be better than even the 8370E if it was better than the fastest FX..

Anyway thank for the caricatural answer, that s urban legends facing real numbers..

 

[Phynaz]

100% fud...

Who is your "we", btw, the usual suspects that constantly bash AMD..?.

Actually 100% true, confirmed by at least MSI. "We" would be most people that understand these things.

You've heard it before and attempted to discredit it before. I'm sure that won't stop you from trying again, even though nobody will believe you this time either.

 

[Abwx]

That is total SYSTEM peak power, it has nothing to do with CPU 125W TDP.

For those that are interested in the CPU TDP under prime 95 here the results at Hardware.fr, CPU power comsumption on the 12V rail is on the right as "ATX 12V (W)".

Losses in the VRMs are included and the number should be factored by 0.9 to get the actual TDP, these numbers are 23% higher than what is yielded with regular MT softs.

At 4GHz with user settings , and with Prime 95 TDP is 108W, at 4.5GHz it is 160W, under regular MT loading those numbers are 20% lower.

I reposted AT measurements as it remarkably show the voltage binnings between a 8320E and the 8370E.

The latter is only 1.21% better at 4.5GHz but its advantage increase to 4.76% at 4.8GHz, overall that s a difference little enough that there s about no point in buying the 8370E in respect of the much cheaper 8320E.

 

[myocardia]

Do you know how to read a graphs i posted..?.

Here the average of thoses tests :

What is this "anything you re talking about"?.

Your i5 should be better than even the 8370E if it was better than the fastest FX..

Anyway thank for the caricatural answer, that s urban legends facing real numbers..

Of course I do, peewee. Intel's fastest 4-core CPU, the 4790k, beats AMDs fastest CPU in pretty much everything, like I said the first time, and your chart here proves. Just let me know when you want some lessons in reading graphs.

 

[Abwx]

Actually 100% true, confirmed by at least MSI. "We" would be most people that understand these things.

You've heard it before and attempted to discredit it before. I'm sure that won't stop you from trying again, even though nobody will believe you this time either.

You should provide numbers and sources, all i see are blank statements, what about the measurements aboves..?.

Or do you consider that these cant be understood by people who understand these things..?.

Or perhaps your point is that you, and some others, have better understanding of electrical laws than me..?..

It is as likely as you having a better understanding of processes than IdontCare, not that he cant be wrong in matters that are not his specialty but personaly i wouldnt challenge him in his domain of excellency, seems that you are not as cautious apparently...

 

[Abwx]

Intel's fastest 4-core CPU, the 4790k, beats AMDs fastest CPU in pretty much everything,

Lol, that s what you call to beat something..?.

10-15%..?.

According to these graphs the 4770K/4790K are CPU that have terrible perf/$ ratio, 10-15% better perfs for 100-200% more money since a FX8320E is equivalent to about any other FX given that it s unlocked.

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

 

[Phynaz]

You should provide numbers and sources, all i see are blank statements, what about the measurements aboves..?.

Or do you consider that these cant be understood by people who understand these things..?.

Or perhaps your point is that you, and some others, have better understanding of electrical laws than me..?..

It is as likely as you having a better understanding of processes than IdontCare, not that he cant be wrong in matters that are not his specialty but personaly i wouldnt challenge him in his domain of excellency, seems that you are not as cautious apparently...

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

 

[Keysplayr]

Lol, that s what you call to beat something..?.

10-15%..?.

According to these graphs the 4770K/4790K are CPU that have terrible perf/$ ratio, 10-15% better perfs for 100-200% more money since a FX8320E is equivalent to about any other FX given that it s unlocked.

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

If it was the other way around, I'm certain you'd feel 0.0005% would be enough. 10-15% would be substantial, yes.

Here's a good question. How much would a buyer have to spend on an Intel CPU, to equal AMD's fastest CPU? That would include motherboard as both platforms could be had fairly cheaply.

 

[Phynaz]

If it was the other way around, I'm certain you'd feel 0.0005% would be enough. 10-15% would be substantial, yes.

Lol, true!

 

[BigDaveX]

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?

 

[Abwx]

If it was the other way around, I'm certain you'd feel 0.0005% would be enough. 10-15% would be substantial, yes.

The thing is that it s not the other way around , so dont put your own thoughts inside the brains of the others as justification of your own bias.

Reality says that it s not the other way around, and if it was it would change nothing, 10-15% better perfs are not worth 100-200% more money.

Here's a good question. How much would a buyer have to spend on an Intel CPU, to equal AMD's fastest CPU? That would include motherboard as both platforms could be had fairly cheaply.

For the price of an i3 you can get a FX8320E, or this latter plus MB for the price of an i5.
The savings doesnt end here, though, a FX8320E is unlocked and as such it s a discounted FX8350.

The usual power argument doesnt hold, difference are small enough that they wont be recoupped by eventual electricity savings.

At worst, for whom is living in the UK, Germany, Illinois or NY, not only there will be no saving at all but it will cost more in the long term, only results (do the maths for Chicago for instance..) will be someone living in the cold to recoup the savings that were actualy highjacked by Intel....

 

[Abwx]

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?

Actualy Northwood wasnt bad at all assuming reasonable frequencies, at 2.8 it was decent, but what you miss is that it had no price advantage even with lower models..

I dont remember Intel pricing them at almost half the price at comparable throughputs than the competition, or was it the case..??.

 

[AtenRa]

Actually 100% true, confirmed by at least MSI

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.

 

[AtenRa]

Here's a good question. How much would a buyer have to spend on an Intel CPU, to equal AMD's fastest CPU? That would include motherboard as both platforms could be had fairly cheaply.

According to this graph

and the latest prices from newegg, you will have to spend $145 for the AMD FX8320E and $310 for the Core i7 4790.

Cheapest Intel socket 1150 mobo (H81) starts at $42

Total = 310 + 42 = $352

AMD 970 that will let you OC to 4.6GHz no questions asked (ASUS M5A97 R2.0) starts at $80

Total 145 + 80 = $225