The New AMD Building Thread

Right on man. That is one sexy motherboard. I would have jumped on it had it been available when I went Phenom II. My DFI board is quite capable though and not worth the extra cost to replace.

 

Indeed LOL! I'll be patient with my board, until its time to upgrade. Im not interested in spending MORE on a new board. 10 Sata ports!!! NOW your talking LOL! Coupled with the 955 could be like a dream ehh? 3.8 with no voltage bump! NICE NICE NICE.

Who knows. Maybe in a few months, I can sell my board, in the hopes of yet a new build. We'll see. My mother is considering an upgrade. Perhaps I can give her the deal of a life time.

 

Sam,
You yourself recently told someone that you can't just set an Intel and an AMD to the same clock speed and test them because they scale differently, and in my opinion, undervolting is just plain cheating.

All you mention is 4 tests! Anandtech ran 24 tests, including some Multi-threaded ones against a wide range of Intel's as well as other AMD's, and when the smoke cleared the AMD 955 had done well enough for Anandtech to say, "The Phenom II X4 955 draws slightly more power than the Core 2 Quad Q9550, but not by a significant amount. Based on its performance, the 955 takes the energy efficiency crown away from the Q9550 in the majority of our tests".

TechSpot says the difference is +11w at idle for the AM3, and +13w for the AM2+. Under load, they show a completely different picture, where the 955 AM3 draws 31w less than the Q9550 under load, and the AM2+ draws 25w less. These were "Retail" chips. the first one's out!

One other thing I would like to ask about Silentpcreview's tests. Where are you getting that 130w figure from?



Please note, that they are rated by their loaded power draw, from lowest to highest. None of the CPUs tested were below 140Cw at idle, and that was the i7 920! I think in the overall scheme of things, there will be very little distance between the 9550/9650 and the 955. In fact since the 955 doesn't draw as much power under load in the real world, they should be fairly equal overall for everyday real world use. It makes AMD more than competitive with the Q9550, mainly because mainstream buyers don't OC, and that's where they money is to be made! The 955 also takes a couple of pot shots at the Q9650 and the i7 920 along the way! Indeed, AMD is back and competitive again, and we still won't have the Full Monty until some time in May! Good stuff for AMD and Intel fans alike! It should keep the prices down!

Best Regards,
Russ

 

130W is a rough approximation. Taking a 75% efficiency figure (PSU + Vreg) off the difference in load gives me 35W. Assuming the Q9650 uses 95W, 95+35=130W.
You'll notice they didn't undervolt the Core 2 to save energy, even though they could have. What difference does it make? Undervolting the AMD lets it do the same job and use less power doing it, I don't see what the problem is.
Also note, the Q9650 is merely in there because it's the CPU they happened to run the test with. The Q9550 results, while simulated, are amost identical to what an actual Q9550 would achieve.
The purpose of these 4 tests is to illustrate that there are 4 different situations where the power consumption was measured, and in each one, the AMD CPU was way behind the Intel power-wise.

I'll pass on reading what Anandtech has to say on this, but SPCR have one of the most thorough power testing methods I know, so I'll take their power results over anyone else's. Also, I think you're belittling the overclocking argument. True, not everybody does it, but I think you'd be surprised to find out just how many people do, when you start buying £200 CPUs.

 

but they are both great OCers, DDR3 costs around the same as DDR2, the mobos are around the same price and same priced CPUs.

ones better in soem tests, the other is better in others.

so the obvious choice is to use a via nano CPU in a netbook.... wait, worng conversation xD

in the end, you should recommend either one, as IMO they are good as each other, and wellcome back AMD , if the perosn prefers intel, recommend that, if they want AMD choose this, now there is no need to recommend the other company.

asi remember incracing the clock speed of AMD CPUs, also increace the speed of the "uncore", and so they scale much better OCed

anyways ill go back under my rock now

 

Agreed, they're both good chips, there's no denying it. Still, for the people that want the most out of a platform for that price, it's still going to be Intel, conditionally on if they want to overclock.

 

but they both OC very well....

 

19-25% versus 28-45%. There's well, and then there's very well.

 

I, for one, am excited to see AMD return to the enthusiast level. I am extremely happy with the performance and overclocking of my Phenom II. I was also able to get 4GHz stable but only with 1.55v which means it gets pretty warm and sucks power like nobody's business, lol.

As far as brand preference, I am definitely an AMD/ATi guy. They offer high performance, decent value, chipsets as good as Intel's, and formidable overclocking. Intel/Nvidia have also released some smashing deals lately and I'm not opposed to using them if the price is right. In fact, my Phenom II was only a side-grade from my Q6600. I only made the switch because I was itching for a good AMD since the failure of Phenom XD

 

Sam,
No offense, but I don't see how you can learn much by passing on a review from a company just because you don't much like them. I don't like SPCR very much either, but I still read their reviews of things. You can say what you like, but that's what's called "Skewing" a test! Especially when Anandtech uses software to test with, that many of us use every day, against a much smaller number of tests. You simply can't prove overall power consumption using that narrow a view! We pay for the power we consume, so why should someone go by 6 tests that include 4 rarely used Apps. Win RAR and ITunes are about the only really used Apps, and Win RAR isn't exactly used that often. I fail to see how those tests demonstrate the overall power efficiency of any CPU! I think it says a lot for AMD, that their efficiency, under load is better than what they were designed to compete against. Not by any huge amount, but still better I look at it this way, last year at this time, we wouldn't have even been having this conversation! LOL!!

AMD has come a long way in the past year, and have overcome their shortcomings to the point of finally being competitive with Intel. True, they can't match Core i7 yet, but the prices will have to drop, particularly for the motherboard, before you can even consider comparing Phenom II to Core i7. People that don't know diddly about computers are not going to spend $100 or more additional money for something they know little or nothing about!

I wouldn't be surprised at all to find out how many people overclock, but in the overall scheme of things, more than 90% of all computers sold are either never, or can't be overclocked, and come with OBG! Most of those people wouldn't know a PSU from a CPU cooler or a video card! Most people are afraid to take the side cover off! I recently went with a Dentist friend to buy some DDR for his office Dells. Even though I argued with him that I could get better memory for a lot less money, he was so reluctant to add them himself, he paid $270 for two no-name 512 sticks from Netseller in Riverside, just so I could install them for him while I was there! Crazy! Like most people, he's scared to death of anything inside inside the computer. My friend Fred (fredbun) was terrified during his entire build, as he had already wiped out his old HP computer just adding some more memory to it! In Fred's defense, the design of the case and the location of the DIMM slots on the motherboard, made it difficult for even an experienced technician to do right. Poor quality memory slots didn't help either. I know because I replaced a motherboard in the same model HP last year. I still can't believe that an intelligent person like this Doctor, would spend $357 for a replacement Propitiatory HP motherboard, and then pay me to change them out. For the $507 he spent, I could have built him a much better computer than an Athlon 3800+ single core! Either a low cost AMD Dual core on a 780GX motherboard or a low cost Intel C2D would have rocked his world, for less money! His comment was, "But it's an HP", like it was something to be "Revered"! Ah, the power of advertising! ROFLMSOAO

Russ

 

you know Russ some people just don't know what there buying these days...hold it I know (store bought) pc's limited to upgrading in some factors...hey I did'nt tell u when...i picked up the Tpower I45 and the Q9400 for my co-worker's build...at micro center the saleman beside the other saleman behind the counter tryed to Con me into putting the cpu on the board and test it for $19.95 extra...I throw some old slang on him about a few programs like sandra and others,,and he did'nt even know nothing of what...i was saying or never heard of it...I blow his mind he was lost...lol it's funny when you go in a computer store like micro center,,etc and pick up a few parts and you know more than the saleman does and he looks at like you should be working there some of the stuff..I see people buy...I fell sorry for cause they don't know what there getting or they don't know how to read reviews or don't care...and saleman there just for a comission sale

 

Will,
Still not a bad idea, for a merchant like Micro-Center. It doesn't have to be fancy or include all the testing we do, but just to be sure if it works properly. It would save a lot of returns!

Russ

 

HAHA LOL!!! When I went to get my 22" widescreen at best buy, the salesman tried to convince me that a 24" would make my computer perform better. I started in about how higher resolution equates to worse performance and how I was buying the 22" display specifically for the purpose of using mid-range video cards, etc. I think I lost him completely when I told him I already had a 24" lol. He was like WTF then why are you buying a smaller monitor? It blew his mind when I told him I had 3 PCs LOL!

Got my 9800GTX today. It's a Zotac 9800GTX+ AMP! Edition. It handles Crysis admirably at 1680 x 1050 in my dual core AMD rig. All high with DX9 "Cheap" Very High minus sunshafts. 23FPS minimum,36FPS average, high of 43FPS on the benchmark. That's the similar to my 8800GTS 512MB on my quad core Intel rig. Except the quad core rig is even using performance mods XD The X2 7750 BE is turning out to be a solid dual core performer. It puts up well against comparable mid range dual core Intels in benchmarks and even plays games like the more expensive dual Intels. It's not even Phenom II based and I'm very surprised with its performance.

 

Estuansis,
I hate to tell you this, but if the Native resolution of that 22" monitor is 1680x1050, it's a terrible choice! They don't seem to make many in higher resolutions. The problem is the increase in the amount of area on the screen that does nothing, is huge. You would be far better off getting one of these, but it's a 21.5 with a very low 0.248mm Pixel Pitch, compared to 0.282mm for the 22" ones!
http://www.newegg.com/Product/Product.aspx?Item=N82E16824236051
They bother my eyes. That 2" stretch for 1680x1050 is just too much. You can get the contrast right and get it all set up, and it just never looks quite sharp enough. It has a soft quality to the picture, that I just don't like, and it's hard to read print on! Pixel Pitch isn't everything, but I've compared one to my Sceptre, with it's slightly better SPVA screen, with the Asus also plugged into my video card, and I want one! I would be perfect for me, as I don't have a very big room. I tried a friends 24", but it totally overwhelmed me! Too big for the space. Games are another story as it's great for games.

Seriously though, If you just bought it and it's resolution is 1680x1050, consider taking it back and get that Asus. The aspect ratio is a perfect 16x9, and the picture quality is crystal clear. I would call the color, "Lush"! It's just a much better monitor, and free shipping makes it a great buy! I want one! Asus also has one that lacks an idle mode to save energy, that you have to shut off manually or the light blinks at you forever, for $10 less!

Both give you VGA and DVI and HDMI cables I already have to turn my monitor off anyway, so I wouldn't have to change anything from what I do now! Also free shipping!

Best Regards,
Russ

 

Lol, a perfect 16x9? I much prefer 16x10 for games as it's more widely supported.
16x10 22" monitors at 1680x1050 look fine. The picture will only go "soft" if they aren't run at their native resolution or if they're run over VGA.

Russ: The SPCR test was never intended for performance measurement, I was using it for more accurate power consumption figures.

 

I use iTunes every day from. The moment I turn on my pc to the moment I shut off, and I use winrar a good few times a day, Sam you got a link to those spcr review so I can check it out?

 

Hmm well, to be fair, how often do you use iTunes in a CPU-intensive manner? It's only adding files and converting WMA to AAC that does it really. Perhaps the visualiser I suppose.

http://www.silentpcreview.com/PhenomII-955

 

Sam,
I think in the overall picture, they are only taking on the highest power usage apps as examples. I fail to see how they can judge overall power consumption that way. As they said in the article, The AMD Phenom II 955 used less watts in the majority of the tests Anandtech ran. I believe there was 26 in all! Based on that statement, and backed by their review and tests, I would guess that there wouldn't be a nickel's worth of difference, between one or the other in everyday use!

Russ

 

ok for a sec lets dismiss SPCR and anandtech, what do other sites such as xbits labs/bit-tech hexus, etc say?

 

Shaff,
Playing iTunes is different! LOL!! I use winrar a lot myself, with most files being 50-60MB and more. I sent Gina the ISO for XP-64 bit that I had winrar split into 7 parts. The whole process took about 2 hours with sendthisfile.com. If I had sent it in one big 588MB file, they would have throttled it so bad it would have taken about 7 hours all together! Either that or the transfer failed at 97% done! This way keeping the parts below 100MB, they barely get throttled at all. I also got free use of another service Gina belongs to. Grasshopper, or something like that. It worked more like File Hosting service's do. In fact, it had been a file hosting service that the Government busted along with about 40 others for porn. After the charges were dropped because they were clean, I guess they decided to use their website in a much different way. They also don't do on-line storage. It sends, it's gone! LOL!! Sendthisfile is faster, but between the two of them, it wasn't bad. I also got to spend a couple of hours with my Gal in Chicago on the computer, yakking with her! She's just come up with some serious life changing health issues, caused by her Diabetes. She used to be around AD as gina3XOXOXO. I built her a computer about 3 years ago! Sweet gal, but way too young at 37, to have to deal with it this!

As far as the SPCR goes, I think their testing for power consumption is terribly flawed! I fail to see any way that you can consider 6 tests that are known to run poorly on an AMD, in terms of wattage consumption. I don't dispute the results, I dispute the conclusion they came to, based on their methodology in testing! What's the point when you only look for known bads? That's what I meant when I told Sam about the tests being Skewed!

Best Regards,
Russ