But it offers less than half the performance? :shrug:
Printable View
Yay! GG again, AMD. 2 good news in one day! :)
Good news as in... mobilize the giant's big lazy butt to make some price reductions. I want octo nehalem for 150$ !
Quote:
Originally Posted by gallardo
Keep on dreaming man...Try 400+$ for lowest clocked 8 Core Nehalem(whenever it comes out).
Waiting for the B3 revision for this, even if the B3 doesn't offer more oc headroom, it's still nice to have the errata fixed.
How do you factor that?
Core 2 is around 10% faster clock-for-clock, that means Q6600 is equal to around a 2.6GHz Phenom. How do you get 1.8GHz being "less than half" of 2.6GHz?
Even overclocking, if we assume this chip can hit 2.4GHz and an average Q6600 will hit 3.6GHz (around 3.9GHz Phenom) it's still far from less than half.
if the northbridge is clocked at 1600mhz, and is stable up to about 2000mhz, does that mean we can expect about 2000/1600 * 1.8ghz = 2.25ghz from it?
more or less?
or am i talking crap
One thing is certain - this CPU will be perfect tool for testing the HTT ceiling of AM2+ mobos!
It's reasonable to expect it can reach 2.4 GHz a speed that is minimum for ALL that have tried Phenom 9500!
For it to reach 2.4 GHz, you'll need to raise reference clock from 200 to 266 MHz.
Regarding the NB, 266x8 would result in 2133MHz, and so far reaching 2.2 GHz for NB didn't prove to be a problem… so yeah this could be really one sweet cheap QC processor!
Wow thats awesome if the price actually does happen. I have a q6600 and its great but for 125 bucks which is 1/2 the price i would totally have gotten this.
OK, I have to ask, do you even play games? If you actually played games you will realise that a 1.8GHz Phenom will bottleneck any modern GPU... in terms of gaming performance it'll be about equal to a stock E2160 or X2 4000+... in other words, it sucks for gaming. A higher clocked dual core for the same price would be much better.
yeah, here are some gaming benchmarks of a phenom quad with cores disabled
http://www.tomshardware.com/2008/02/...nom/page5.html
over 2 cores there's virtually no improvement except for supreme commander
that'd have a very trivial impact on the core-scaling in games, the overall picture is pretty clear
We made a review
9100e Vs C2D4700
It seems like the Entry Quad Core AMD is just marketing around the word "Quad"
The review is in Greek, but if you see the graphs...you will understand everything i guess...
http://www.hwbox.gr/showthread.php?t=1219
ouch, loosing against a e4700 sure hurts (which is known to oc like crazy and produces far less heat)...
Amd should just hire the apple marketing guys, then they could sell a 500 mhz quadcore with tlb errata for 1000$:p:
The performance results from the Greek review are brutal.
and isnt there gonna be special pricing per 1,000 chips or something like that??
e7400 GMV for my next CPU...
http://www.hwbox.gr/showthread.php?t=1219&garpg=8 (credit to GoriLLakoS for the review).
As I said earlier, higher clocked dual core >>> lower clocked quad when it comes to gaming.
If you bothered to take a look at Winrar results for 9100e you would see it had TLB fix turned on...Testing this CPU with the patch on brings the results further down(apart from low clock speed and singlethreaded nature of a lot of tests done).
I can't believe it took 20+ posts for somebody to mention overclocking one.
I heard that the 9100e was already available in an HP (i think), and that if you can find one for sale, itll cost you about as much as a 9500 would.
I'm a bit baffled by those results. There seems to be something wrong with your setup. It looks like you most likely have the TLB patch enabled. Also i didnt see any mention whether you ran the tests with memory in ganged or unganged mode. Iirc there is a new version of both Sandra and Everest that added proper support for Phenom. Did you use those? Additionally you used mostly single threaded apps where obviously(granted not so obviously for ppl that dont know much about computer technology) the faster clocked processor is going to be faster. Power consumption results are also not directly comparable since you werent able to use CnQ.
p.s. I'm trying to be polite here.