You do that quite fine on your own...I wasn't refering to a DC DDR beating RDRAM, I was refering to your repeated claims over the past few months(the last time I was here a couple months ago you were saying much the same thing as now)that SC DDR solutions beat (they squeak past PC800 if at all)1066 which isn't true. Oh, BTW it's DUAL <-- channel, not DUEL(unless the sticks of memory are fighting to the death).
XP2200+ for $152? God damnit I could have got one of those instead of my 1.8a. Then I wouldn't have to worry about overclocking not so good.
Oh well my chip shuld reach 2.4 easily (I hope) with some decent air cooling and a ducting system.
Oh well my chip shuld reach 2.4 easily (I hope) with some decent air cooling and a ducting system.
Originally posted by: CrazySaint
Actually, I didn't know that, mostly because to the best of my knowledge, its not true. Check the benchmarks in this review at AT. Near as I can tell, there's not a huge difference between DDR, PC800, and PC1066. Certainly not enough to have a bearing on the PR ratings.
And I pointed out that whatever they may use as the basis of their PR rating, it appears to be pretty fair.
Got any links?
Actually, I did, and it looks impressive. IIRC, I beat PC1066 by about 3-5%.
Until I see proof that HT increases performance by over 20%, I still say that 30% between HT and GB is optimistic.
Dude, chill. I'm not anybody's case. This is a technology forum, we're discussing technology.
yodayoda said:
here is a parting shot. do you know that old joke about discussion forums?: how is winning an argument in a discussion forum like winning the Special Olympics? either way, SSXeon5 is retarded.
Man, grow up. We're talking about freakin' cars and technology here, insulting each other over it is just stupid.
HERE is where the i845G out performs the i850 w/ PC-800 a large amount of times. And GREAT!!!! you saw the Granite bay review before they took it down lol, yeah it did very well for a alpha board and cas2.5
Old Preview on HT is a great way to see how apps can benefit and not benefit from HT. For instance in 3dstudio Max 4.26 there is a 17% increase with a single HT cpu enabled, and Lightwave 7B(ray tracing) there is a 18% increase! And the 10-30% is mostly in server apps, like Microsoft SQL Server, and Microsoft IIS.
Sorry bout lashing at you Crazysaint but im sick of these personal attacks and constant flamage.
SSXeon
Interesting review. Yeah, the DDR beat the PC800 one many of the benchmarks, but it also trailed the PC800 on quite a few. THG doesn't seem happy unless they run AT LEAST 2,934 benchmarks for each review (
), so I didn't bother counting benches to see which RAM type was a head in most benchmarks, but it appeared pretty even. Also, near as I could tell, when the PC2700 DID lead the PC800, it was usually by 1-3%, and I don't think it was every by more than 5% (and that, rarely). I would probably, call it a tie, or perhaps give the DDR a very slight lead, but certainly nothing that would warrant calling a system running PC800 "handicapped". In all honesty, I would imagine that if you setup three identical P4 systems side-by-side, with the only difference being that one system uses PC800, one PC2700, and the third PC1066, I'm guessing that with the possible exception of a few specific apps, you would be very hard pressed to tell a difference between them. And darn you for making me "defend" RDRAM! And GREAT!!!! you saw the Granite bay review before they took it down lol, yeah it did very well for a alpha board and cas2.5
Yep, very nice
Btw, did you read Anand's article on Intel's 0.9 process? Very impressive! Well, first off that article talks about Xeons, not "ordinary" P4s (although a Northwood should presumably have similar results in the same apps), second, as you said, by far, the most largest gains are in server apps, which has very little to do with what most people use - the same can be said for 3DSMax. Also, there were several benchmarks where enabling HT actually caused a decrease in performance, including Photoshop, which is the closest benchmark in that article that covers "real world" performance for most people. So, to say that enabling HT automatically offers a "30% increase" in performance without any caveats, while technically accurate, is a bit misleading to most people. I could just as easily say it causes a 19% drop in performance.
No problem, some people just get too worked up about these things.
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
I just want a fast computer that runs my apps at light speed, or close to it. I have owned AMD and Intel and been very happy with either.
My current Choice is AMD, this may change, but SSxeon, you are an Intel Fanboy to the max who started this thread to bang on about PR rating and to sh1t all over AMD. If you really care so much about absolute screaming power, why do you only run an old P3 @ just over 1Ghz?
Why, for example, do you not run a Xeon....
So do not be surprised when ppl get upset, when you know that is exactly what you wanted. Oh, and this isn't your thread, it's our thread. You're just the member who started it.
My current Choice is AMD, this may change, but SSxeon, you are an Intel Fanboy to the max who started this thread to bang on about PR rating and to sh1t all over AMD. If you really care so much about absolute screaming power, why do you only run an old P3 @ just over 1Ghz?
Why, for example, do you not run a Xeon....
So do not be surprised when ppl get upset, when you know that is exactly what you wanted. Oh, and this isn't your thread, it's our thread. You're just the member who started it.
Once again just dont post, why did you need to say that? Crazysaint ended the thread nicely and I wanted it to just die, Like I sain 100000 times, i started it to see if it was right that the hammer would have to be at 2.6Ghz to accoplish a 3400+ PR raiting. Then I did state how I dislike the PR raiting. Why do I still have a p3, have you seen my others specs, 955DF ($300) Radeon 8500 64MB (got it 2nd day of launch $280) 100MB Zip ($38) Logitech Optical Wireless Mouse ($50). So i kinda got things that I needed at that point instead of a P4 upgrade when the prices were still a bit high. Im planing to grab a 2.53/i845G/PC2700/radeon 9700 combo in the next few months.
SSXeon
Originally posted by: CrazySaint
Interesting review. Yeah, the DDR beat the PC800 one many of the benchmarks, but it also trailed the PC800 on quite a few. THG doesn't seem happy unless they run AT LEAST 2,934 benchmarks for each review (
Yep, very nice
Well, first off that article talks about Xeons, not "ordinary" P4s (although a Northwood should presumably have similar results in the same apps), second, as you said, by far, the most largest gains are in server apps, which has very little to do with what most people use - the same can be said for 3DSMax. Also, there were several benchmarks where enabling HT actually caused a decrease in performance, including Photoshop, which is the closest benchmark in that article that covers "real world" performance for most people. So, to say that enabling HT automatically offers a "30% increase" in performance without any caveats, while technically accurate, is a bit misleading to most people. I could just as easily say it causes a 19% drop in performance.
No problem, some people just get too worked up about these things.
Yeah, but I do like toms reviews he gets them out sometimes before anandtech and I hate waiting for reviews im so impatient
Your right, I shouldnt call a PC800 system handicaped, but I still HATE RDRAM with a passion. That .09um actical is very very nice and pretty much a nice follow up to there previous one. The thing I guess I failed to mention is to check out the date on that HT preview, its very old and like SSE2 got optimised, so will HT in the coming monthes. I also mentioned that there is a decrease in some tests, namly my beloved photoshop But I really hope this will change, photoshop is heavly optimised for SMP and HT is not far off SSXeon
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
You were spouting about how you believe in free speech in this very thread and you have the nerve to try to gag me? Hypocrite...
No you come in and pick at me, like everyone here, and then when i flame back, BLOODYMURDER!! so dont try to call me a hypocrite. You start to talk about how my computer isnt fast, im an intel fanboy and then say that all i wanted is people to get upset. hmmm
SSXeon
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
Originally posted by: SSXeon5
No you come in and pick at me, like everyone here, and then when i flame back, BLOODYMURDER!! so dont try to call me a hypocrite. You start to talk about how my computer isnt fast, im an intel fanboy and then say that all i wanted is people to get upset. hmmm
SSXeon
You told me not to post, you advocate free speech, therfore you are a HYPOCRITE.
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
Originally posted by: SSXeon5
You just dont get it, STFU.
SSXeon
I should do that for You? You don't get it. I don't take orders from Peons like yourself.
Foo.
Originally posted by: DivideBYZero
Originally posted by: SSXeon5
You just dont get it, STFU.
SSXeon
I should do that for You? You don't get it. I don't take orders from Peons like yourself.
Foo.
GO HERE AND STFU
SSXeon
actually dividebyzero you talk pretty big for a newcomer to these forums, nothing you said had any bearing on the thread what so ever, so please post elsewhere.
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
LOL! Your 455 posts really impress me. Go see NSF4's count and come back to talk about n00bs.
My addition to the thread was related to it flame atrracting subject. If you follow SSXeons posts you will no doubt have noticed that he is a rancid Intel Fanboy/AMD hater who trolls for flame.
And before we get the '*WHINE* But your an AMD Fanboy *WHINE*' read the previous posts in this thread.
DivideBYZero
Lifer
- May 18, 2001
- 24,117
- 2
- 0
Originally posted by: SSXeon5
Originally posted by: DivideBYZero
Originally posted by: SSXeon5
You just dont get it, STFU.
SSXeon
I should do that for You? You don't get it. I don't take orders from Peons like yourself.
Foo.
GO HERE AND STFU
SSXeon
Is that your latent homosexuality coming out? LMAO!
Yeesh guys, grow up already. Insulting each other over hardware is just stupid. This thread is starting to sound like a video game chat room.
SSxeon5 seems to have a rather intelligent argument most of the time. He seems accurate for the most part when sharing his opinion about processors.
True, he usually posts good stuff. It usually sparks some interesting debate and sometimes some flame wars. He just needs to be a little more thick-skinned about the thread crapping...whining (okay, griping) about it, then blaming product zealots for all his woes, just isn't becoming at all.
I'm not a huge fan of the PR system, but here is why I do see decent justification for its existance ... I'll use vehicular engine analogy since there are a bunch of car buffs here.
I'm using a very simplified model of a processor performance, FREQUENCY * IPC = processor performance as analogous (in a philosophical way) to RPM * TORQUE = power. FREQUENCY and RPM in a way represents the # of events occurring in a given amount of time, and IPC and TORQUE represent the amount of "work" that can be done per event. Of course, in a car, measuring power is relatively straight forward, and at the very least is shown to be exact theoretically. There is no theoretical bases for something like this in a processor (as of yet anyways). However, back to the analogy ... imagine if people sold performance vehicles based on RPM. So, Honda comes out with a "new" high performance 9000 RPM engine and says look, we have the highest RPM engine! Well, the Corvette (stock anyways) does not rev to 9000 RPM, so does it make it a "weaker" car/engine? Well, since the Honda has peak power at 8300 rpm, it's torque is only about 152 ft-lbs. The Corvette, on the other hand, has a torque of 355 ft-lbs. SO, while it's true the Honda can achieve higher RPMs than the Corvette, the Corvette has a much higher overall power. So, a cheesy "PR"version for a Corvette would be like a "13993+" S2000
The best solution would be to have an "unbiased" performance metric - but in the processor world, that would be impossible.
BTW, a food for thought. Where would the world (or just the US hehe) be without "high performance" race cars? It'll be less fun, but I don't see too much change in society. Now, where would the world be without heavy duty diesel engines ... LOL, it'll be really screwed up! How do you think the really nice INTEL and AMD cpu, groceries, mail, ... practically everything in life, get from the factory to the retail store? It's not just trucks, but many ships and trains use diesel engines as well. AND, how well do you think all these really cool exotic cars, mentioned in this thread, will do trying to pull a 10,000 lb trailor (not to mention 80,000+lb trailor)? And do it with 15,000 - 65,000 mile oil drain intervals, a 300,000+ mile no major engine problem even when the power of the car and the truck is the same! So, just like in cars and computers (and everything else - Sir Mix-A-Lot comes to mind too hehe) there is not such thing as one size fits all (or one benchmark).
Peace
I'm using a very simplified model of a processor performance, FREQUENCY * IPC = processor performance as analogous (in a philosophical way) to RPM * TORQUE = power. FREQUENCY and RPM in a way represents the # of events occurring in a given amount of time, and IPC and TORQUE represent the amount of "work" that can be done per event. Of course, in a car, measuring power is relatively straight forward, and at the very least is shown to be exact theoretically. There is no theoretical bases for something like this in a processor (as of yet anyways). However, back to the analogy ... imagine if people sold performance vehicles based on RPM. So, Honda comes out with a "new" high performance 9000 RPM engine and says look, we have the highest RPM engine! Well, the Corvette (stock anyways) does not rev to 9000 RPM, so does it make it a "weaker" car/engine? Well, since the Honda has peak power at 8300 rpm, it's torque is only about 152 ft-lbs. The Corvette, on the other hand, has a torque of 355 ft-lbs. SO, while it's true the Honda can achieve higher RPMs than the Corvette, the Corvette has a much higher overall power. So, a cheesy "PR"version for a Corvette would be like a "13993+" S2000
The best solution would be to have an "unbiased" performance metric - but in the processor world, that would be impossible.
BTW, a food for thought. Where would the world (or just the US hehe) be without "high performance" race cars? It'll be less fun, but I don't see too much change in society. Now, where would the world be without heavy duty diesel engines ... LOL, it'll be really screwed up! How do you think the really nice INTEL and AMD cpu, groceries, mail, ... practically everything in life, get from the factory to the retail store? It's not just trucks, but many ships and trains use diesel engines as well. AND, how well do you think all these really cool exotic cars, mentioned in this thread, will do trying to pull a 10,000 lb trailor (not to mention 80,000+lb trailor)? And do it with 15,000 - 65,000 mile oil drain intervals, a 300,000+ mile no major engine problem
even when the power of the car and the truck is the same! So, just like in cars and computers (and everything else - Sir Mix-A-Lot comes to mind too hehe) there is not such thing as one size fits all (or one benchmark).Peace
wow, thats one of the better posts i've seen on this forum in a while. and oh *GASP* it didnt come from someone with 18million posts. just 1 post (atleast right now). yet again proving you dont have to be here forever to be smart.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 24K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 20K
-
-
-
Facebook
Twitter