• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Anandtech Haswell review is up!

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I don't really care about that galego, AMD isn't a factor in my price range or for my needs.

If you want to start an Intel vs AMD war do it in another thread.
 
galego said:
Do you believe that TR selecting the more power hungry (Asus) mobo for FX and the most power saving (MSI) mobo for i7 when compared the power consumptions of AMD vs Intel chips was... arbitrary? No.

A change in motherboard can give 20-30 W for the same chip same everything else.
Click to expand...
Way to pull that one way out of left field.
 
pretty disappointing that it can get so hot, and that OC for the non K is a lot worse than before.

apart from that, GT2 is an OK improvement, GT3e is to expensive...

Ivy Bridge is looking even nicer now...
 
I wouldn't mind stepping up to a Haswell CPU, thinking of it since I apparently need another desktop for the house and wouldn't mind upgrading to a nicer rig and retiring this to someone else.

That being said I'd only really be willing if some OC enthusiasts get under the IHS and tell me what's there, the Ivy temps could have been so much better if they didn't switch to whatever it is they put under there vs previous generations.
 
WhoBeDaPlaya said:
Need bottom line answer to :

1) Whether non-K SKUs will support BCLK OCing (wording in AT review is a little vague)
2) Whether K SKUs will have TSX
Click to expand...

Just noticed this excerpt from Xbitlabs :

"All crucial internal frequencies remain at their defaults in this case, but the x86 cores, the uncore part, the integrated graphics core and system memory get overclocked proportionally. It also means that even LGA1150 CPUs with locked multiplier can be overclocked, but only by 25% or 66% above the default frequency."
 
jpiniero said:
Wonder if Intel raised the TDP more to save money by selling "leakier" parts for the 84W models.
Click to expand...

When you widen the entire cache stucture to 256bit from 128bit and double the speed, not to mention beefier IGP and IVR. You are bound to use more.
 
IntelUser2000 said:
Applies all in partially loaded situations, and at same TDP, which isn't what a 84W 4770K is versus a 77W 3770K.
Click to expand...
Real world tests suggest a greater power delta, however. 7W TDP difference is rather misleading. Haswell isn't power efficient past certain Ghz level, at all.

Idle power is nice, but it isn't as important to me.
 
Last edited:
WhoBeDaPlaya said:
Just noticed this excerpt from Xbitlabs :

"All crucial internal frequencies remain at their defaults in this case, but the x86 cores, the uncore part, the integrated graphics core and system memory get overclocked proportionally. It also means that even LGA1150 CPUs with locked multiplier can be overclocked, but only by 25% or 66% above the default frequency."
Click to expand...

this is wrong, hardware.fr had a locked haswell i5 and the bclk divider thing is locked, so you only have around 7% of OC by using the BCLK, exactly like on ivy bridge, and to make things worse, the Turbo OC (which could add 500-400MHz on ivy/sandy bridge) is also more limited now...
so for non K CPUs things got a lot worse.

http://www.hardware.fr/articles/897-6/overclocking-plus-libre-k-plus-strict-par-ailleurs.html

(use google translator)
 
http://www.anandtech.com/show/6993/intel-iris-pro-5200-graphics-review-core-i74950hq-tested/4

The Crystalwell die measures 7mm x 12mm (84mm^2), while the quad-core Haswell + GT3 die is a whopping 264mm^2 (16.2mm x 16.3mm). Working backwards from the official data Intel provided (177mm^2 for quad-core GT2), I came up with an 87mm^2 adder for the extra hardware in Haswell GT3 vs. GT2. Doubling that 87mm^2 we get a rough idea of how big the full 40 EU Haswell GPU might be: 174mm^2. If my math is right, this means that in a quad-core Haswell GT3 die, around 65% of the die area is GPU. This is contrary to the ~33% in a quad-core Haswell GT2. I suspect a dual-core + GT3 design is at least half GPU.
Click to expand...
1: Quad core GT3(core i7 4770R) die size is enormous for a 22nm process, at 32nm it would be close to 400+mm2, much like an Okta-core 39xx SB-E. Since Intel will price the GT3 at $350-400 and make a nice profit keeping high margins, ill imagine they can do the same and introduce a nice 8-core 22nm die(250mm2 ???) and price a 6 core CPU at $300-350.
For those of you(and me) that will expect something like that forget it, they will not introduce a 6 core at the same price as Core i7 4770K 🙄

2: If the iGPU part of the Quad Core GT3 is 65% of the entire die, then the iGPU of the Dual Core will be even higher at 75-80% and not half(50%).

3: I believe it is time for AMD to release a mobile 300mm2 SR Quad Core, 768 GCN cores Kaveri 28nm at 55W TDP (+GDDR5) and sell it for $400. :sneaky:
 
Abwx said:
Benchmarks are non neutral numbers...

When Intel agreed to help MS optimize Excel we can be sure
that the Intel devellopper put their opinion about the "best way"
to "optimize" this soft , so ultimately the rzesults are influenced
by the devellopper opinions , we are not in pure mathemlatics
where results are independant of humans wills...
Click to expand...

What's with
the random line breaks?
I always picture
you composing your posts
under a cherry Tree wearing
victorian costume while wracked with
Consumption.
The content is usually
interesting but the
format is challenging.

😉
 
WelshBloke said:
What's with
the random line breaks?
I always picture
you composing your posts
under a cherry Tree wearing
victorian costume while wracked with
Consumption.
The content is usually
interesting but the
format is challenging.

😉
Click to expand...

Some crap browser on a non-win system is my guess...
 
monstercameron said:
i'm not sure that is a fair assessment, the amd a10-5800k(ddr3 2133) still beats the hd4600(with ddr3 2400), however crystalwell is an equalizer which does. Hypothetical, if the a10-5800k had as much bandwidth do you think gt3e would still beat? in any case it does and amd needs to step up their game...
Click to expand...


Yes it should. GT3e has more computing power.
 
lol @ forum fail on haswell article being posted. I can only imagine the rabid fanboys desperate to post their positive/negative spins on it. 😉

For me it's meh as I have a 2500k with a decent overclock.

Can someone tell me what situations/workloads AVX2 etc will make a difference in? I mainly only game on this system.
 
At least the battery life improvements on the mobile Haswell CPUs will be quite ok though:

http://www.engadget.com/2013/06/01/intel-haswell-launch/

The 50% improvement that Intel previously has claimed only seems to be during video playback though. MobileMark2012 gives 6.1h (IB 3667U) -> 8.3h (Haswell 4650U) on a 50 Wh battery, which equates to 36% improvement.

Noway near the perf/watt improvements we saw going from AMD Zacate->Kabini though. To be fair that was moving to both a new process node and uarch though.
 
AtenRa said:
http://www.anandtech.com/show/6993/intel-iris-pro-5200-graphics-review-core-i74950hq-tested/4
2: If the iGPU part of the Quad Core GT3 is 65% of the entire die, then the iGPU of the Dual Core will be even higher at 75-80% and not half(50%).

3: I believe it is time for AMD to release a mobile 300mm2 SR Quad Core, 768 GCN cores Kaveri 28nm at 55W TDP (+GDDR5) and sell it for $400. :sneaky:
Click to expand...

One of the poster said it was surprising for Intel to surpass AMD on iGPU so quickly, its not surprising when they dedicate such massive die space for their iGPU, and on a node advantage.

But the fact is, if Iris Pro is ~165mm2 or so, that's within range of a decent dedicated GPU die, which it has no hope of matching.

All that adds to the $$.
 
Silverforce11 said:
One of the poster said it was surprising for Intel to surpass AMD on iGPU so quickly, its not surprising when they dedicate such massive die space for their iGPU, and on a node advantage.

But the fact is, if Iris Pro is ~165mm2 or so, that's within range of a decent dedicated GPU die, which it has no hope of matching.

All that adds to the $$.
Click to expand...

Yes, just to show you how inefficient(performance per die size) the Intel Iris is, at almost the same die size of 160mm2, AMDs HD7790 (including 128bit memory controller) should be 50% faster or more at 28nm process.
 
AtenRa said:
Yes, just to show you how inefficient(performance per die size) the Intel Iris is, at almost the same die size of 160mm2, AMDs HD7790 (including 128bit memory controller) should be 50% faster or more at 28nm process.
Click to expand...

How much power does the 7790 use again? 10-15W? Noo...it uses what? 85W TDP? 😉

I wonder what you gonna promote next with BS to make Intel look bad and AMD look good...

I've already vacationed one person for thread crapping; I don't want to make it two. This needs to stop right now.
-ViRGE
 
Last edited by a moderator:
ShintaiDK said:
I wonder what you gonna promote next with BS to make Intel look bad and AMD look good...
Click to expand...

Im not the one produced a 174mm2(for the GT3 iGPU) + 87mm2 for the eDRAM die that perform lower than the 118mm2 GTX650M (including the 128bit memory controller) 😉
 
ShintaiDK said:
Most likely around 50W.

Whats next? First it was 7790. Now its 7750? How far down are we gonna go? SPs are already reduced from 896 to 512...
Click to expand...

Just to show you that even with 512 cores (even smaller die size of 123mm2) it would still be way faster than Iris. :sneaky:
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top