Whatever... guess no proof...Done with this conversation...I link numbers and all you can do is claim BS but cannot prove it....
Whatever... guess no proof...Done with this conversation...I link numbers and all you can do is claim BS but cannot prove it....
Lihan, maybe if you link us to the numbers you claim, you would be in a better position, but as Duvie said , you claim something you can't prove.
AMD Venice 3200+ Idle at 5-10watts, less with CnQ enabled. Total system power usage for a stripped down, single drive, 1gb Amd system runs in the 110-140watt area (more if taken at the wall). Intel compairisons work out similarly. So that means that a bare system draws at least 100watts typically. If you have sli'd video cards, as many of these testers do, a raided pair of drives, an independant sound card, an optical drive, and a few case fans, you're looking at 135-175watts easily, before the cpu is taken into account.From there on it's simple subtraction and division.
Carlosd, I took the numbers directly from the tests he himself linked.
Now if you want it to get really complex and damning to his arguement further, you can calculate the accuracy of Intels TDP ratings versus AMD's TDP ratings by looking at Amd's TDP and actual tested power draws (For reference AMD claims that they rate for the highest power consumption that their cpus can ever physically have when run at specified speed and voltage, intel rates at 75% of that value). Then you can look at system power numbers, deduce the maximum possible draw that could be from the cpu, and by that determine the system's power usage and isolate the cpu usage.
But it's much simpler to take the test which has the best testing method of the two he posted, adjust it to accomidate the 35% innaccuracy he blames on their failing to fully load the processor and calculate it out to 1:3.7 as I have done.
yeah!!! :laugh::roll: G o ahead then and use the reviewer setup and break it all down and show me how it works.....You still wont get 1/4th to 1/5th as you originally claimed...The best you are going to get is the 1/2.5 that my other links show.....Just face it you came up a bit short....
I list 2 articles that break it down to just cpu power and even show more then just the 820D.....Why dont you show how they screwed up their data...Then go tell them>>>Some how I think they are a bit above you....
Well the Xbitlabs the 820 shows, 1/2.5
Game PC 1:1.5 for the 840ee (ok they show total system consumption, an it could be not so accurate)
Lost circuits 1:2.7
I don't get your point
Whatever the ratio is, the PD is a very hot CPU.
Game PC 1:1.5 for the 840ee (ok they show total system consumption, an it could be not so accurate)
Lost circuits 1:2.7
I don't get your point
Whatever the ratio is, the PD is a very hot CPU.
You are the one who claims that they weren't fully loading their cpus. Which in case you do not realize it (as you clearly don't) is an admition that their results are skewed to show a significantly low load draw in compairison to an accurate idle draw, which adjusts the ratio to the 2.5:1 you claim. You have admitted that 2.5:1 can't be accurate because their load draw was read significantly short of full load, yet you still insist that it is accurate. You contridict yourself. Whereas, following your own statement that the load draw was read low, I performed a simple calculation which proves that using your own data, the ratio achieved is 3.7:1.
Carlosd, the tests disagree with eachother. There is a 35% difference between Xbits results and Lost Circuits results. Duvie blames that on his assertion that Lost Circuit failed to load their cpus properly. Since this is all directly proportionate, we can estimate that if you raise Lost Circuit's load results by 35%, then you will achieve accurate readings, assuming that Xbit WAS fully loading their processors. This is all a direct conversion. There is absolutely no possibility of error outside of Duvie's claims. It's as simple a compairison as you can possibly have. And it proves absolutely, that if Duvie was correct, and both tests are accurate to the extent HE decided. The Answer must be 1:3.7.
I'm sorry, but that is as simple as it can be put. It requires a little bit of reading comprehension to follow, but it's not so difficult that I should have required half a dozen posts to explain it. If you don't get it now, frankly, it's because you aren't smart enough to understand what I'm saying. You may disagree with my results, but as I said, the only possible place for error in that calculation is the data Duvie himself presented. The math is flawless.
I'm sorry, but that is as simple as it can be put. It requires a little bit of reading comprehension to follow, but it's not so difficult that I should have required half a dozen posts to explain it. If you don't get it now, frankly, it's because you aren't smart enough to understand what I'm saying. You may disagree with my results, but as I said, the only possible place for error in that calculation is the data Duvie himself presented. The math is flawless.
So lihan show us a link showing clearly 1/5 ratio you claim. Anyway the ratio thing is not important, but the numbers itself, who cares if the CPU go 50 W idle (who cares idle?) when it reachs those 200W you claim full load, it is HOT, and that is the importan issue.
n7
Elite Member
- Jan 4, 2004
- 21,281
- 4
- 81
Can you guys act any more childish?
We all know P-Ds have high power consumption issues regardless of whether they are idling or are under load.
:roll:
Also, while i do have a rather strong bias against Asus (i had two supposedly-great Asus mobos that i hated & had issues with), the fact is, it's not a good deal.
I'd get the MSI Neo4-F rather than the Asus A8N-E, since it's $30 Cdn. cheaper, & has everything the Asus does, & likely will be better for OCing if needed.
We all know P-Ds have high power consumption issues regardless of whether they are idling or are under load.
:roll:
Also, while i do have a rather strong bias against Asus (i had two supposedly-great Asus mobos that i hated & had issues with), the fact is, it's not a good deal.
I'd get the MSI Neo4-F rather than the Asus A8N-E, since it's $30 Cdn. cheaper, & has everything the Asus does, & likely will be better for OCing if needed.
I actually said ~1/4th to 1/5th and was refering to Intel procs, not dual cores specifically. Since dual cores are likely the worst case scenario for this, then I think 1/3.7th was close enough to accurate that the ~ justified my statement. And I'm not trying to say that Intel's are cool processors. But In the Athlon XP days, judging from temps I saw, processors were using close to 50% power when Idle, It wasn't until Venice really where the A64 surpassed the pentiums in power savings while idle. Because of this, a processor that uses just over 25% power when Idle, I feel justified labeling as cool while Idling.
Power savings, as said ratios are not important, but the numbers itself, since you are comparing CPUs with totally diferent fabrication process.
Markfw
Moderator Emeritus, Elite Member
- May 16, 2002
- 27,331
- 16,162
- 136
Lithan, keeping the slide rule out of it for the moment, all three sites you linked to make it apparant, the P4 and the P-D processors take a LOT more power than the X2 at idle OR full load, and they increase a LOT more from idle to full power. All three reviews show that, but the way they got there differs, and thus the numbers differ.
Give it up man !!!
Give it up man !!!
Mark, this has nothing to do with compairing the power usage of x2's and p4d's. If you read the thread through, you'll see that it's about how a P4 duallie shouldn't be running 85*C @ idle. I made a post that explained why that was and duvie went on an attack rampage against me. So I proved my point, which was mostly that he was talking out of his ass because he's a fanboy. Hell, I wouldn't keep a p4-D if it were given to me. I hate the damn things, but suggesting that it isn't peculiar for them to idle at unsafe temperatures as Duvie did is quite frankly, stupid.
Markfw
Moderator Emeritus, Elite Member
- May 16, 2002
- 27,331
- 16,162
- 136
I have one. They are a wierd animal. And I also said that I haven;t seen those temps, but I could actually believe that a motherboard could report that, especially if shutdown was disabled. Forget the the85c argument as thats NOT what you have spent most of your time arguing, you have been trying to prove this 1/4th to 1/5th power change from idle to load on Intel, and that doesn;t hold water, and thats what is being argued.
Again, admit when you are wrong Lithan, grow up.
Again, admit when you are wrong Lithan, grow up.
Actually I though you was triying to prove the 1/4, 1/5 argument. Duvie was just laughing about your statemet that the PDs idle very low. Very low compared to what? It idles toghether the 670 higher than any other CPU.
Very low compaired to what they run at when they aren't idling.
And I've got some tbirds and pally's that would disagree with you there. I've seen idling in the 60's on stock cooling with these. The cooling is important because it's relative. A cooler that's designed to run a 150watt proc (perhaps even overclocked beyond that) that's max temp is ~70*C @ load shouldn't be idling @ 85*C. Especially when the chips do increase heat output by such a large amount when loaded. Now my buddies 1900+ Palimino would idle @ 60, but still only ran ~ 72*C @ load. A 820D that idles at 60*C is going to be boiling water before it hits full load.
And I've got some tbirds and pally's that would disagree with you there. I've seen idling in the 60's on stock cooling with these. The cooling is important because it's relative. A cooler that's designed to run a 150watt proc (perhaps even overclocked beyond that) that's max temp is ~70*C @ load shouldn't be idling @ 85*C. Especially when the chips do increase heat output by such a large amount when loaded. Now my buddies 1900+ Palimino would idle @ 60, but still only ran ~ 72*C @ load. A 820D that idles at 60*C is going to be boiling water before it hits full load.
Markfw
Moderator Emeritus, Elite Member
- May 16, 2002
- 27,331
- 16,162
- 136
80 watts less compared to 30 watts less for the X2 ? You can't spin the numbers any way, they suck a lot of power anytime ! jeesh, give it up !!!!
50 watts less at idle vs 100 watts less at load than the X2.. No matter which way you cut it the phrase "less power" and Pentium don't go together.
And don;t mince numbers, they just aren;t in the same ballpark. (I'm tired and going to bed)
50 watts less at idle vs 100 watts less at load than the X2.. No matter which way you cut it the phrase "less power" and Pentium don't go together.
And don;t mince numbers, they just aren;t in the same ballpark. (I'm tired and going to bed)
Mark just stop typing until you've actually read what I said. I compair processors idling abilities based on the efficiency itself, NOT on the processors overall efficiency. A processor that runs at 300*C at load and only 75*C at idle is very efficient at idling despite the fact that this other processor runs cooler. The issue was a processor running much hotter than it should while idling. It was not that the processor was too hot to cool. It seems it was just an ineffective (passive) cooler. You can take up your grudge with the p-d's temp problems with someone else because I agree with you there. But they run much cooler when idling than when loaded, and I'd never seen a loaded p-d running at the temps he was running at Idle, so I thought something was wrong because I had missed where he mentioned that it was a passive heatsink. So I was attempting to explain to him that Idling @ 85*C was not normal, and then on cue the amd fanboys come in and start giving me crap because this and that benchmark says that they run 1/13th hotter than I had said, as an approximation, that they should run at idle.
And once I put them into a corner using their own claims about the data, where they should have to admit either lying or being wrong. They immediately start saying, "Well that doesn't matter because amd still runs cooler", as though I had ever said anything to the contrary.
And once I put them into a corner using their own claims about the data, where they should have to admit either lying or being wrong. They immediately start saying, "Well that doesn't matter because amd still runs cooler", as though I had ever said anything to the contrary.
lol, you have some nerve insulting someone's rig, especially when you don't know anything about it or how it performs, why don't you get your facts straight? and who still buys retail? maybe those whom buy Dells or what not, but this is AnandTech, the home of the Ferengis and Hot Deals. i wouldn't even have considered a P-D until i went searching out of curiousity of how cheap it was. (OT: tthe secret of my 'cheap' P-D was the search string: nobody would have searched 820D, most people only search 820 or *D 820). now i'm scouring ebay for the X2 3800 because i can't believe someone did a BIN for $230.
are you new or something? talk of the price of the p-d has been beaten to death. look earlier in this thread (around post 40-50) when that idiot fatty kept budding his stupid @$$ in, i had to prove to him how retarded his "arguments" were.
i would have been sure that you would have seen the thread when i first got my 820 rig and detailing the choices of why i got the cpu, and especially with the motherboard (so i didn't have to upgrade to PCI-E, and i had a choice of DDR and DDR2. ocworkbench seemed to favor the board, and i knew it wasn't going to be o/cing very well, since it had no Vcore adjust. but a 12.5% o/c wasn't too bad (it almost gave me an 840), and i still kept all my 'old' components.
after a brief discussion on whether to choose regular ddr 400 or fast ddr2 667,
i ended up doing the autoGK bench w/ DDR and DDR2, ,with 1h28min for DDR-400 2-3-2-6 and 1h30min for DDR2-667 4-4-4-12 (considered LL). And for some reason, when I ran it again with Crucial ValueSelect, my new score is 1hr27min DDR-400 2.5-3-3-8.
the point is, i'm just trying to be more realistic about things. my 'cheap' p-d is only good now, until i have to pay the electricity bills. but at least during the winter, i don't have to turn up the heater in my room.
325USD/550USD = about 59% of the X2 in price (which will only become less over time). but you know what, that's also about the same performance difference, too. i only try to tell it straight, i don't BS anything, and showed that (sadly), cheap DDR400 == cheap DDRII667. the motherboard certainly doesn't seem crappy (only in the o/cing dept), but its serving its purpose well. i wanted a cheap dual core system, and that's what i got (i do run multi-threaded apps, so its not a waste of $$$). i'm just waiting on a super loco deal on an X2...
I had the chance a few weeks ago to use my brothers computer that has an AMD X2 4200 and to say the least I was impressed. It also had a gig of ram and the 6200 turbo cache from GeForce and it ran nicely. Not exactly the best gaming rig but I really noticed a HUGE step up from what I was running at home (in my profile).
Instead of the smoke and mirrors act... why don't you discredit him since what your saying IMPLIES that you have read both.
Personnaly I would NEVER buy a MSI product but that is just me.
To the OP:
If you want a good system then make sure you by an ABIT board whether you go for the Intel D or Athlon X2. Personnaly I would go for the X2 since the P-D will make up the difference in cost at the end of the month (light bill) for the duration of its lifetime and then some.
Okay, so what am I supposed to think here?
(all prices CDN$)
Asus A8N-E - $132
MSI K8N Neo4-F - $102.99
Abit AN8 Ultra - $155.25 (at another place, as the one where I got prices for the other two doesn't have any socket 939 Abits)
Should I be starting a thread in the mobo forum asking for suggestions on those?
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter