• 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.

Haswell Benchmarks?

D

dbcoopernz

Member
Over at Xbit Labs they're reposting some alleged benchmarks.

http://www.xbitlabs.com/news/cpu/di...tel_Core_i_Haswell_Processor_Hit_the_Web.html

OClab.ru, a Russian web-site dedicated to overclocking, has published what it claims to be the first test results of an engineering sample of quad-core Intel Haswell processor clocked at 2.80GHz. The chip was tested in SuperPi 1M&32M, PiFast, wPrime 32M&1024M on an unknown mainboard under Microsoft Windows 7 64-bit operating system. The obtained benchmark results could not be verified and it is unknown whether they are accurate.
Click to expand...


x5Qb85S.png



jWVzFB6.png
 
Test system?

http://oclab.ru/news/pervyie-rezult...tsessora-haswell-v-populyarnyih-prilozheniyah

Процессор INTEL Haswell 2.8 ГГц, Intel Core i7-3770K@2.8 ГГц
Память Corsair Dominator 2666МГц@ 1333 МГц CL9-9-9-27 Блок питания Corsair АX1200 Вт
SSD Corsair Neutron 240 Гбайт
Операционная система Windows 7 x64
 
Go to Who's blog . He says any of these early results are not the results that he is getting in the lab . WHO KNOWS!We have all seen that picture befor
 
So I should maybe skip haswell and wait for the next generation before I upgrade my 2500k? 😕😕😕
 
No. Well maybe... are you happy with your 2500K? Does it bottleneck?

I wouldn't expect Haswell to blow Ivy into irrelevance (or Sandy for that matter) but it should be a little better than the increase Ivy was to Sandy... the main benefit would be overclocking... BLCK + MLT and hopefully none of this de-lidding nonsense.

So... Ivy-E + 5%...

Post-sclaimer: Obviously most of what I said is bulls**t speculation.
 
Well, we'll see. Not a good thing if Intel wants superior mobile performance, since mobile products' "clockspeed wall" for x86 processors is below 3 GHz. I am skeptical of the results, but then again it is only Super Pi, a rather limited benchmark.
 
Torn Mind said:
Well, we'll see. Not a good thing if Intel wants superior mobile performance, since mobile products' "clockspeed wall" for x86 processors is below 3 GHz. I am skeptical of the results, but then again it is only Super Pi, a rather limited benchmark.
Click to expand...

Disappointing results if true, but who knows? I was hoping for 10% performance gain clock for clock vs Ivy. Overall the results given show on average almost no gain.
 
Torn Mind said:
Well, we'll see. Not a good thing if Intel wants superior mobile performance, since mobile products' "clockspeed wall" for x86 processors is below 3 GHz. I am skeptical of the results, but then again it is only Super Pi, a rather limited benchmark.
Click to expand...

ASSUMING these are legit (a big assumption), we still don't have another critical part of the equation, energy consumption.

If this is at 95W, yes they are not impressive. If they are at 5W, then those results are amazing.
 
Vectronic said:
No. Well maybe... are you happy with your 2500K? Does it bottleneck?

I wouldn't expect Haswell to blow Ivy into irrelevance (or Sandy for that matter) but it should be a little better than the increase Ivy was to Sandy... the main benefit would be overclocking... BLCK + MLT and hopefully none of this de-lidding nonsense.

So... Ivy-E + 5%...

Post-sclaimer: Obviously most of what I said is bulls**t speculation.
Click to expand...

I only do light gaming 1080p on my PC and most of the time I just use it to watch youtube or chatting online. The few games I play right now are diablo 3, starcraft 2 (buying sc2 expansion when release in march) and maybe little bit of GTA IV. I was thinking about getting haswell and radeon 8850 (sea island) when release later in June this year. Guess I'll wait. 😉
 
podspi said:
ASSUMING these are legit (a big assumption), we still don't have another critical part of the equation, energy consumption.

If this is at 95W, yes they are not impressive. If they are at 5W, then those results are amazing.
Click to expand...

I don't really expect the power consumption will be much lower, if lower at all than Sandy Bridge. That being said, I also doubt it will be much higher. I would like to see some more benchmarks though (especially ones that are confirmed) before deciding on the worth of this processor. I admit that the results of these alleged benchmarks aren't exactly living up to expectations, but they aren't bad. Nothing really worth spending extra cash on unless they really shine somewhere.
 
Not exactly a credible source, but even if it were true, I wouldn't really care.

What I want from Haswell is lower power consumption and better graphics, single threaded performance is fine as it is.
 
Well, knowing that intel is focusing in performance/watt, this results may be true. I dont expect too much of an IPC jump, but maybe a good gain in multithreading (thanks to HT, as a single CPU core has much more resources now).
 
Realworldtech said SB was supposed to 10% faster clock-for-clock than Nehalem which ended up being correct and now they said the same for Haswell vs SB.
 
I knew it... Haswell is just a big GPU upgrade and power consumption. Ill grab myself a Ivy Bridge E in Q3 2013.
 
StrangerGuy said:
Realworldtech said SB was supposed to 10% faster clock-for-clock than Nehalem which ended up being correct and now they said the same for Haswell vs SB.
Click to expand...

Most of the enhancements are ones they did on Sandy.

Enhanced pre-decode.

Sandy added the uop cache in decode, Haswell doesn't enhance decode besides the combined buffer afterward.

Enhanced OoO engine but less enhanced than SB over Nehalem.

The biggest thing is the added integer unit and store unit. They also gave cache doubled bandwidth, but that's possibly for AVX2 use.
---
Since this is an enhanced Core, I would be surprised to see it backpedal anywhere.
 
podspi said:
ASSUMING these are legit (a big assumption), we still don't have another critical part of the equation, energy consumption.

If this is at 95W, yes they are not impressive. If they are at 5W, then those results are amazing.
Click to expand...
Same process,more complex chip and that much better power draw? No way. It may draw ~10% less tho,if this was the case than it has a good perf./watt improvement(even with this kind of small IPC increase,for a tock that is).
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top