Ever since I heard of these 2 chips recently, I've been wondering how these would stack up against the Phenom II X4 905e. I'm currently using a 4850e in my HTPC and have been considering the 905e as an upgrade. The 2 chips reviewed here have higher frequencies but also higher stock TDP and no L3 cache. Considering I can source a 630 @ ~140 CDN and a 905e @ ~195 CDN (~40% premium), I'm wondering how close they are in efficiency and performance? I'm still leaning to the 905e but am now a little torn.

I've seen tests comparing power consumption of 905e and 620 but only at stock voltage. 905e was superior but couldn't match Intel's quads unfortunately.

Will SPCR review the upcoming 45W quads from AMD? I would like to see a power efficient quad roundup.

Hi,

was this undervolted through BIOS or software? Didn't manage to catch any explanation on how the processors were unvolted.

Anantech has a new tool which will allow you to compare the 4850e with the 910, which as close as you're going to get with their list of cpu's, which is pretty close.

Mike, one thing you didn't mention in the article (it was very good) was how the memory divider circuit works on the AMD processors. Anything other than an even whole multiplier may not be running your memory as fast as you think. Here's the 2 cpus covered in this article.

The 630 at 2.8GHz runs DDR2-800 at it's full speed. Here's the calculation, using 400MHz for the memory, the real speed not DDR2 speed:

2800 (CPU) divided by 400 (memory) = 7.000. There is no remainder, so the divider does not round the number up. So the divider remains at 7 and we have 2800 divided by 7 (the divider) = 400 or full speed.

The 2600 on the other hand is not as good. 2600 divided by 400 = 6.5. The memory divider can't handle fractions and will not overclock the memory, so the divider it chooses here is also 7. So 2600 divided by 7 = 371. So your DDR2-800 memory is actually running like DDR2-742.

The 4850e, a 2.5 GHz model, has the same issue. 2500 / 400 = 6.25. It also gets rounded to 7. 2500 / 7 = 357, or DDR2-714. You can confirm all this with CPU-Z. I run my 4850e at 2.4GHz, so that memory will run at full speed.

The memory divider doesn't take OCing into the calculation so when you up the FSB you up everything, but you still need to know where you start.

Mike, the UV settings seem right in line with the "e" series of CPUs. They run at 2.5GHz with a tiny amount above 1.2 volts, setting a larger safety margin voltage-wise than your findings, but probably a good idea to ensure everything off the assembly line runs without the need for individual testing and settings. You didn't mention if the Q8200S runs at reduced voltages, did you?

I applaud Mike's UV testing. What the people who do full time folding are slowly finding out is that undervolting often leads to far more efficient use of power, which is a big deal when you pc runs under 100% load 24/7.

We used AMD Overdrive to determine the undervoltage (it's a lot quicker than BIOS because it doesn't require rebooting). However Overdrive disables C&Q, so once we found the minimum voltage, we set it in the BIOS, then ran our tests.

We actually did test the 905e, but since it's closer to $200 than $150, we left it off the charts. I can add the numbers later if you want, but it was roughly equivalent to the undervolted X4 620. Definitely not worth the extra money.

I see your X4 630 has AACYC markings, which indicates that it is possibly based on a Deneb core with L3 cache disabled. Have you had any success "unlocking" it using ACC option in BIOS?

It would be good to see the Lynnfield data added to the chart also. I appreciate that the price is higher but pricing is only one factor when building a system and it’s close enough to the others to warrant inclusion IMO. I consider more expensive CPUs if they offer better performance, lower power consumption and good over-clocking/under-volting potential.

These new Athlons sound good and there’s so much choice currently and when Clarkdale is released it might also nip at the heals of low end Quads which opens up even more choice in the $100 - $200 segment.

Can you please provide a link? I was on Anandtech.com, but couldnt find that tool...

I'm also running a 4850e and for my purposes it's generally good. But when I edit photos.... So I'm very interested in the new 620. But how much faster will it be?

Thanks, Thomas

EDIT: Found it here: http://anandtech.com/bench/default.aspx?p=36&p2=85

Given the 620 ~ the 910... it appears the 620 will be approx twice as fast, as the 4850e in PhotoShop / photo editing. Quite good for such a low price, me thinks

I can accept that and look past the difference in TDP of 65w and 95w as the slight savings in energy would take forever to make up for the difference in price; especially if they perform work at the same rate. Now I'm waffling between the 620 and the 630 however. The 630 seems to perform ~7% better than the 620 but also uses ~7% more energy (~10% when both are UV'd) for ~27% more in cash as I can source a 620 at ~$110 CDN. That basically leads me to believe that they can both perform the same amount of work for the same amount of power; only the 630 accomplishes it's work faster. I can live with the slower 620 to save the cash unless I'm overlooking something that makes the 630 more compelling.

620 vs. 630, well, with the pricing where I live, it's an easy choise: The 630 is ~50% more expensive, and I wont pay that for 7% extra performance.

I don't understand the need to overprice items in Europe. The 630 in the US is only 20% higher in price. Does this price include a VAT?



If we go to this link and click on the CPUZ screen on the right we see that the reviewers at the French website got very similar results undervolting their 620.

http://www.pcinpact.com/actu/news/53087 ... ervolt.htm

Maybe Mike reads French?

VAT is included - meanwhile prices have changed, and now the difference is approx. 20%. - the 630 price have dropped

When some other AMD processors was introduced within this year, prices also changed within the first week or two. But back then, it was the cheap model, which got more expensive...

I read french - but dont understand much

Fortunately, babelfish does: http://dk.babelfish.yahoo.com/translate ... ers%C3%A6t

Babelfish hurts...

I'm not a translator, but I live and work in english and french every day, so here's my quick-and-dirty translation of the pcimpact.com article:

"As we have seen in our Athlon II X4 620 preview, AMD has once again opted to go with a relatively high voltage despite the reduced size of the die and a frequency of "only" 2.6GHz.

The VID displayed by our motherboards is 1.4V and the voltage is 1.12V at idle and 1.392V at load. The total measured power draw on our test machine ran between 99 and 178 watts.

Since we're not really fans of wasted watts, we once again tried to reduce the processor voltage to it's lowest possible levels while still keeping a stable system.

To do this, we used our GA-MA790FXT-UD5P motherboard which allows us to reduce the CPU voltage from the BIOS.

After many tries, we managed to get a stable system at 2.6GHz with a voltage of 1.15V, a reduction of 17.8%. This resulted in a voltage at idle 0.864V and 1.152V at load.

Once again, we would have liked AMD to have made a different choice for the voltage. Most users do not undervolt and it's critical that manufacturers set more reasonable values.

Hopefully the motherboard manufacturers are listening, and change their aim for future products."

Like I say, I'm not a professional translator, but hopefully this has been easier to read than the Babelfish version.

Cheers,

-Scot

I am a long time SPCR reader, and occational poster. Recently I bought X4 620 along with ASUS M4A785-M Motherboard and installation process.

I saw that you undervolted the X4 620 to 1.1625 and system used about 124 watts at full CPU load.

Even for my CPU that is how far I can undervolt, but my system is consuming much less power.

Idle about 54 watts which is inline with yours( adding 10watts VGA card idle power)
CPU full load(all cores Max power) is of about 98watts.

Ofcourse I am using onboard video( Which should about 10w less when GPU is idle based on the video card power consumption) and OEM heatsink.

Here is my config

ASUS M4A785-M
X4 620
2G DDR2 800
LG Blu-Ray/HD-DVD Drive
SEAGATE 7200RPM HDD
120MM Case Fan
Antec Earthwatts 430W power supply

But one thing I observed, it seems despite 1.1625v core voltage setting in Bios, vCore is fluctuating between 1.136v to 1.15v. But it seems giving it little bit more 1.675v did not change the power consumption(may be 1 watt which is fluctuating anyways).

So for me I have full quadcore PC under 100watt at full CPU load, and idling below 55watts with Normal Desktop HDD and Optical Drive. (less than my Q6600 system undervolted)

Is it that the Motherboard you used very in-efficient to almost 15-18watts additional?

Also, one observation, the OEM CPU fan supplied has very rough noise, like as if the bearing are shot. Is that how OEM fan sounds or mine has problem?

chinna_n --

It's difficult to compare your CPU to ours on different mobos, onboard video vs discrete VGA, and different PSU. There are just too many variables. You might also have an exceptionally cool running CPU -- this does happen sometimes.