Early Ivy Bridge Sample Tested

[Eldonko]

Whoo hoo upgrade to a slower PI time! Oh wait I already have that with SB-E

[Hornet331]

I wonder what Paulbagz will make of this, he was so sure that Ivybridge was going to be the second coming.

Everyone who said that had no clue. IB is just a shrink of SB, the preliminary benchmarks are where I expected them. Haswell on the other hand... hmm

[andressergio]

Whoo hoo upgrade to a slower PI time! Oh wait I already have that with SB-E

lmao

[paulbagz]

I wonder what Paulbagz will make of this, he was so sure that Ivybridge was going to be the second coming.

Rofl umad?

I never said anything of the such. I merely commented that people were making comments without any form of benchmarks.

Was never pro/con for IB, just stated people, like you, need to wait for results.

Por que?

-PB

[ice_chill]

It's a bit like when Core 2 was shrank from 65nm to 45nm, nothing big but still a little architecture optimization.

[Brice MJ]

>10% IPC improvement in some tasks?

http://img440.imageshack.us/img440/6830/intelivy.png

[boxleitnerb]

No, mostly turbo I guess.

[paulbagz]

>10% IPC improvement in some tasks?

http://img440.imageshack.us/img440/6830/intelivy.png

So there is >10% increase in IPC (according to that slide)?

-PB

[Brice MJ]

No, mostly turbo I guess.

2600 runs at 3.4/3.8ghz while i7 3770 will be 3.4/3.8ghz, same default clocks
maybe the i7 3770 has an extra bin (+100mhz) with all cores active, but they're pretty close

[RPGWiZaRD]

As usual from die-shrink "tock" at this point u can add the typical 4 ~ 10% IPC increase (around 15% in some special cases) but what is attractive is its performance/power consumption/overclocking. I pray it'll do like 5.5GHz @ 1.35v on an average sample and consume quite a bit less power than Sandy would do at like 5GHz 1.4v. This is just speculation but one can always hope right.

I like throwing my overclock speculation pre-launch so here's mine:

Worst sample: 5.3GHz @ ~1.35v
Average sample: 5.5GHz @ ~1.35v
Best sample: 5.7GHz @ 1.36~1.37v
i5-3750K / i7-3770K on good aircooling

For 24/7 I probably wouldn't use more volts than that on 22nm (with air/water at least).

[Aberration]

There is a lot of work on the manufacturing end that needs to happen for this not to slip another quarter. Also the problems they are having, could lead to very wide ranges of potential across chips.

[Eldonko]

I wish I shared your optimism lol. Im betting on oc the same as SB, the same mhz cap for sub zero, and less than 5% improvement in most tasks.

As usual from die-shrink "tock" at this point u can add the typical 4 ~ 10% IPC increase (around 15% in some special cases) but what is attractive is its performance/power consumption/overclocking. I pray it'll do like 5.5GHz @ 1.35v on an average sample and consume quite a bit less power than Sandy would do at like 5GHz 1.4v. This is just speculation but one can always hope right.

I like throwing my overclock speculation pre-launch so here's mine:

Worst sample: 5.3GHz @ ~1.35v
Average sample: 5.5GHz @ ~1.35v
Best sample: 5.7GHz @ 1.36~1.37v
i5-3750K / i7-3770K on good aircooling

For 24/7 I probably wouldn't use more volts than that on 22nm (with air/water at least).

[Aberration]

It really depends on how the '3D' transistor works ok, and the tweaks that are involved. In reality it is much more than just a normal shrink.

[Philip_J_Fry]

lol when i saw the title i kinda guess its from coolaler... that guy lives next to gigabyte, asus's headquarter or something, and has friends inside.

performance seem ok, for ES. more interested in how much more clock u can get out of it compare to SB.

[RPGWiZaRD]

performance seem ok, for ES. more interested in how much more clock u can get out of it compare to SB.

This, hopefully coolaler can give us an early sneak peak, doubt it'll happen quite yet though, the launch is so far away.

EDIT: Funny how he got like 19s something at 2GHz but 14s something at 2.6GHz with a newer bios, seems like the bios helped quite a bit as well, such a clock doesn't usually result in such increase, so looks like bios optimization has a road to go still. :P

[kl0012]

Don't know if it was already posted, but there are some interesting slides here:
http://hwbot.org/forum/showthread.php?t=34548

[Boogerlad]

that ipc boost seems a bit large...

[Dimitriman]

nothing spectacular, the usual 0-10% gain, but the best thing in IB is the very nice drop in power consumption.

http://www.xtremesystems.org/forums/...re-Goes-Mobile
If the above romours are true then people should get used to not seeing spectacular things from Intel anymore. Expect for spectacularly high prices!

[RPGWiZaRD]

that ipc boost seems a bit large...

Yea 17% sounds a little optimistic (and the first test probably is using AVX) but when isn't a score produced by the manufacturer itself always a bit higher than what average joe gets in the same test? I bet we are more likely to see maybe around 10% or so in those tests. But OMG @ the internal GPU performance increase... sounds like it might be able to play a few even newer games at lower res quite fine.

[Tenknics]

funnny when sb-e was pending everyone was screaming IVY IVY.

Now this leaks and said people are like "what I never said anything about ivy!"

[dartaz]

Yea 17% sounds a little optimistic (and the first test probably is using AVX) but when isn't a score produced by the manufacturer itself always a bit higher than what average joe gets in the same test? I bet we are more likely to see maybe around 10% or so in those tests. But OMG @ the internal GPU performance increase... sounds like it might be able to play a few even newer games at lower res quite fine.

The GPU performance increase is huge because they are comparing HD4000 to HD2000

If they compare it to HD3000 then the increase would be something like 1.5x (which is still impressive)

[blindbox]

Y'all got some high expectations, I'm happy if it'll do 5.5GHz+. Then again I don't have a SB yet. I could use some CPU upgrade though, bloody Skyrim needs a 5.5GHz Sandy Bridge (no joke) to run constant 60 FPS (at least with "Ultra" shadow quality) but I can run 8xAA/16xAF and multisampling nicely with a GTX 460. Looks like CPUs going to become more and more important part for gaming with these shoddy console ports eh?

I've always wondered why on some games, high quality shadow takes a chunk off your processor when old games like F.E.A.R. could do it very well, perfectly.

[STaRGaZeR]

I've always wondered why on some games, high quality shadow takes a chunk off your processor when old games like F.E.A.R. could do it very well, perfectly.

F.E.A.R. took a chunk off your GPU in that case thou

[PatRaceTin]

Damn i5-3570K is my next CPU

[N19h7m4r3]

F.E.A.R. took a chunk off your GPU in that case thou

I really don't mind if it goes about half half really

Although I do like it when games actually use a cpu and it's core propperly