AMD Athlon II X4 620 2.6 GHz

[Bl4ck]

in short : cheap ass quad core cpu , it does exactly that what it supposed to do in it's price range. only thing that is bad it's the power draw ,too much for me.

 

[1Kurgan1]

Very very impressive, I was looking at the non-gaming charts where this thing was snapping at the 965 BE's heals and really starting to wonder, then the gaming tests finally showed up. Still this thing looks amazing for the price, I love the 240 I picked up, and wouldn't mind one of these either.

 

[Zubasa]

in short : cheap ass quad core cpu , it does exactly that what it supposed to do in it's price range. only thing that is bad it's the power draw ,too much for me.

Just under volt the chips if you want low power draw, I can tell you it most likely require less than 1.28V.

 

[1Kurgan1]

in short : cheap ass quad core cpu , it does exactly that what it supposed to do in it's price range. only thing that is bad it's the power draw ,too much for me.

AMD chips always draw more power, but their chipsets draw less, in the end the results are pretty much the same.

 

[Omega]

Power consumption in review refers to TOTAL SYSTEM power draw. Same DDR3 memory is used, same HDD, cooler, etc. So P45+Q8200 draws less power than Athlon II X4 620 + 790FX

 

[1Kurgan1]

The difference is the P45 chipset is not an enthusiast chipset and the wattage difference is negligible, but the 620 does also run significantly cooler. It's hard to compare a p45 and a 790FX, as 790FX has been the top end AMD standard for 2 generations and still has boards that can be up to $200. I'm not completely sure about p45 at the start, but I can't imagine they ever cracked $100 by a lot.

790GX or 785G would be a better comparison, especially with the built in video since thats what a lot of people will be doing with this chip, be a perfect low power HDPC when you cut out a videocard.

 

[ASRockIQ]

just bought this CPU with Rush Shipment via Newegg @11:23AM! Let's see what my system can do now

 

[WarEagleAU]

Thanks Newtekie1 and Omega for answering. I was dumfounded to say the least.

 

[lism]

AMD chips always draw more power, but their chipsets draw less, in the end the results are pretty much the same.

Not really. In practical the power usage is lower then the TDP is. I.e mostly the CPU won't be using 95W at full load. Not even peakening at 95W. Even tho for 4 cores thats low anyway.

Really good price performance ratio. I was thinking to buy another X2 but this X4 really is getting affordable
Edit: Tomshardware doing another review between the same CPU with and without L3 cache. http://www.tomshardware.com/reviews/athlon-l3-cache,2416.html

 

[Omega]

Edit: Tomshardware doing another review between the same CPU with and without L3 cache. http://www.tomshardware.com/reviews/athlon-l3-cache,2416.html

You have that in TPU review already. Please note that i have downclocked Phenom II X4 965 from stock 3.4 GHz to Athlon II X4 620 frequency, 2.6 GHz to simulate performance drop caused by lack of L3 cache.

 

[mastrdrver]

Man, I must have really been asleep yesterday when I saw this article. I missed not only the downclocked 965, but also the PCMark scores.

Hardware Secrets went about comparing this processor in a very different way that I hadn't seen. Instead of comparing against an Intel C2Q or similar price, they compared it to a Intel system of comparable costs. Seeing as the board and cpu are the only real difference, it ended up as a slaughter of a lonely Pentium E6300 except for 2 benches. Not to mention the joke that was gaming with the Intel on board gpu that resulted.

Needless to say, when it comes to either system cost or comparable cpu cost, the Athlon II X4 620 is a worthy foe. It will be interesting to see how a similar priced i5 cpu/system would fair once the dual core/thread i5s show up. It just might get real ugly then for AMD.

 

[MN12BIRD]

Basically it's a Phenom II x4 without any L3 cache. In fact I believe some of them will be exactly that from the factory. But my buddy bought one and for games you usually can't tell its crippled. Well that's what he says anyway! He said most games don't use much cache since they aren't "predictive" and that for most "everyday" programs it seems to be pretty darn quick for a "$100 quad core! I'm sure there are some specific programs that will show its weakness but I'm not sure exactly what they would be. Production work like sound and video editing perhaps??

 

[1Kurgan1]

You have that in TPU review already. Please note that i have downclocked Phenom II X4 965 from stock 3.4 GHz to Athlon II X4 620 frequency, 2.6 GHz to simulate performance drop caused by lack of L3 cache.

Ah I didn't catch that off the bat, still impressive 620, makes me happier now though, I was wondering how their results were so close.

 

[Omega]

Try to read more text guys, everything was explained before the graphs started
Quote from review, page 2:

To find out how much of performance hit removing L3 cache will have, you find a fully potent Phenom II X4 965 in our benchmarks, but clocked down to 2.60 GHz, same as Athlon II X4 620. This will result in an apples-to-apples comparison to easily assess the impact of the missing L3 cache.

Also, in overclocking results, Athlon II X4 620 @ 3.63 is compared to stock frequency Phenom II X4 965 @ 3.4 GHz

 

[I see SPY!]

Try to read more text guys, everything was explained before the graphs started
Quote from review, page 2:


Also, in overclocking results, Athlon II X4 620 @ 3.63 is compared to stock frequency Phenom II X4 965 @ 3.4 GHz

So it seems the cache doesn't impact that much the results...

 

[Omega]

Most of the performance drop is noticed in games, yes. But also note that Athlon II X4 doesn't scale that well with overclocking, as the Phenom II X4 965 has no issues keeping up with the 3.63 GHz Athlon II X4, even if it's clocked 230 MHz lower.

 

[heky]

Omega, why not overclock the two Intel processors to the same frequency as the AMD ones? It would be really interesting to see the clock per clock difference.

 

[W1zzard]

thats not really the point of a review is it ? performance out of the box for the every day user is important, not some technology investigation looking at clock-to-clock performance

 

[Omega]

↑
what he said

When reviewing a product, in this case a CPU, primary focus is performance that product has on factory settings, which 99% of users will use.
Overclocking is just additional content that explores the product potential and to satisfy more experienced users (overclockers).

When Intel processors are tested, they will undergo the same treatment.
Testing clock per clock performance and stuff like that can be saved for occasional non standard review. Will give it some thought in future if things work out

 

[sLowEnd]

Wait, so the IHS isn't soldered on Regors? (referring to pic on p.2 of the review)

 

[Zubasa]

Wait, so the IHS isn't soldered on Regors? (referring to pic on p.2 of the review)

He rip it off.

 

[pantherx12]

Seems like AMD could save a hell of a lot of money by phsycally reduces the L3 cache on later models of Phenom CPUs.

 

[Omega]

Seems like AMD could save a hell of a lot of money by phsycally reduces the L3 cache on later models of Phenom CPUs.

You are forgetting Opteron processors that share the same architecture as Phenoms including L3 cache. And in server applications, L3 cache does maka a whole lot of difference.

In terms of production costs, it more effective for AMD to build just one or two different cores rather than design each specific core for one segment. They don't have the production capacities like Intel to afford such luxury.

Besides, think of the L3 like long term investment. Future six core (twelve core?) CPU's are also going to have the same 6MB L3 cache, and with adding more cores that need more instructions to process, extra buffer really could come handy in some apps.

 

[handsomerichguy]

best price/performance processor ever, I think.

 

[handsomerichguy]

I worry AMD will suffer financial crisis for selling this chip very cheap