Well lets face it RDRAM especially PC1066 has more memory bandwidth than DDR. The Intel 850 series is currently the only RDRAM chipset for the P4. All the benches I see for p4's anymore use PC1066 which is not supported offically by the chipset. All the benches I see for the AthlonXP use DDR333 and the VIA KT333. Now there are many DDR systems shipped by major manufactures with AthlonXP + DDR combos and to my knowledge there are few if any shipping RAMBUS PC1066 + p4. Most P4 systems sold use DDR and not RDRAM. Why should we benchmark P4 + RDRAM vs AthlonXP + DDR? Shouldnt it be AthlonXP + DDR vs. P4 + DDR?
Is it fair to benchmark P4+RDRAM vs AthlonXP+DDR?
- Thread starter majewski9
- Start date
The bulk of all P4 systems will not use RDRAM. All future Athlon systems will use DDR. Chipset makers, the server market, and PC manufacturers have all pretty much denouced RDRAM. Thats why so many DDR chipsets for the p4 are out there. Why cant benches between AthlonXP and 845E+p4 become the mainstay? It just seems like an unfair advantage!
Ideally, a comparison would be between two similarly priced solutions or solutions that use similar hardware setups. A lot of people don't see it that way but if you use two machines that are very similarly priced, I think that would be a fair comparison.
EDIT* No that doesn't mean use RDRAM in the P4 but make up the price difference in the Athlon with obscure, but expensive extras
It means, if I use 512MB of RAM in my Athlon machine, I should use 512MB of RAM in my P4 but I want to make sure it doesnt cost much different..so I just use the same RAM.
EDIT* No that doesn't mean use RDRAM in the P4 but make up the price difference in the Athlon with obscure, but expensive extras
It means, if I use 512MB of RAM in my Athlon machine, I should use 512MB of RAM in my P4 but I want to make sure it doesnt cost much different..so I just use the same RAM.I would like to see both included. After all RDRAM is still pretty popular among enthusiasts.
Let's turn this around a little. Dual channel DDR is coming for the P4. Do you think it will be fair for all reviews to compare this future P4 with dual DDR vs Athlon?
Let's turn this around a little. Dual channel DDR is coming for the P4. Do you think it will be fair for all reviews to compare this future P4 with dual DDR vs Athlon?
"Fair" is meaningless in a benchmark.
These reviews and comparisons are becoming way to spoon-fed. It's like most people see an P4 vs AMD review, jump to the 3dmark and quake numbers, and say that the loser is 0wn3rz3d by the winner. The losers say the benchmark isn't fair.
The web sites like Tom's, Anand's, HardOCP, etc... generally do a fair job of providing data on the performance of different platforms. It's not up to them to decide if a test is "fair".
You have the data. You know how much the chips, mobo's, and memory costs. That's it! Who cares if a website says that the P4 2.8 is faster than an AMD XP Whatever+, your the one who has to shell out the cash if you want to buy it.
These reviews and comparisons are becoming way to spoon-fed. It's like most people see an P4 vs AMD review, jump to the 3dmark and quake numbers, and say that the loser is 0wn3rz3d by the winner. The losers say the benchmark isn't fair.
The web sites like Tom's, Anand's, HardOCP, etc... generally do a fair job of providing data on the performance of different platforms. It's not up to them to decide if a test is "fair".
You have the data. You know how much the chips, mobo's, and memory costs. That's it! Who cares if a website says that the P4 2.8 is faster than an AMD XP Whatever+, your the one who has to shell out the cash if you want to buy it.
I think it's fair as long as you run both IN SPEC, that is, no PC-1066 RDR, nor any 166 DDR MHz FSB Athlons.
But PC800 P4 vs 133/166 MHz DDR FSB/MEM Athlon is fine with me, as would PC1066 RDR P4 vs 166/166 FSB/MEM Athlon be.
But if you're not making an overclocking comparison, no I don't think PC1066 RDR is fair, cause you are in fact running the system out of spec.
But PC800 P4 vs 133/166 MHz DDR FSB/MEM Athlon is fine with me, as would PC1066 RDR P4 vs 166/166 FSB/MEM Athlon be.
But if you're not making an overclocking comparison, no I don't think PC1066 RDR is fair, cause you are in fact running the system out of spec.
The way I see it, it's not only fair but necessary to benchmark a variety of different hardware. I'm not a big fan of benchmarking out of spec/overclocked hardware personally, as much as anything because I'd prefer to see the time used doing more benchmarks of other legitimate if not so exotic things like the P4 with SDRam. Is it slower? Oh yeah, knew upfront it would be, how much though is the question. That particular one was done earlier this year on this site and I for one appreciated the time being taken to compare all the different options (SDRam/DDR/RDRAM)
As far as it being fair to compare the P4 with RDRAM to an Athlon with DDR, as each is the fastest current solution for the platform, it's fair. Then again comparing both with DDR is fair too. To a guy looking to build a system who is planning to use DDR if he goes with a P4 then that comparison is best for him. To someone looking for maximum memory bandwith, the other.
What it comes down to is there are so many options and only so much time and article space for a review, decisions have to be made as to what to include. With any comparison someone can cry foul, that's just part of the fun of having so many options
As far as it being fair to compare the P4 with RDRAM to an Athlon with DDR, as each is the fastest current solution for the platform, it's fair. Then again comparing both with DDR is fair too. To a guy looking to build a system who is planning to use DDR if he goes with a P4 then that comparison is best for him. To someone looking for maximum memory bandwith, the other.
What it comes down to is there are so many options and only so much time and article space for a review, decisions have to be made as to what to include. With any comparison someone can cry foul, that's just part of the fun of having so many options
First off I dont really like the set of benches that Ace's used. Mostly gaming benhcmarks where we know favor the wide bus of the P4. So I dismiss those as being trivial since they really dont test what I use my machine for. Anyway the difference in many game benches are only a couple frames. Who can honestly tell the difference between 103 frames and 107? No one! Then they have some 3d studio max apps which the AthlonXP wins hand over fist, but then we get to the SSE2 benches which are absurd seing how AthlonXP doesnt use SSE2 as of yet.
Also Aces hardware again runs the p4 out of spec by using DDR333 which is not offically supported by the Intel 845G. Also they use an external graphics chip which the home buyer with an 845G will probaly use the intergrated graphics. I hate to be nitty gritty!
Also Aces hardware again runs the p4 out of spec by using DDR333 which is not offically supported by the Intel 845G. Also they use an external graphics chip which the home buyer with an 845G will probaly use the intergrated graphics. I hate to be nitty gritty!
Lol Maj... *A mother's voice to an infant* "U feel better if a benchmark show AMD better?" Get a grip, until AMD starts making mainstream chipsets for AMD's own processors, AMD is going to get owned time and time again....
"Is it fair to benchmark P4+RDRAM vs AthlonXP+DDR?"
Is it fair to benchmark a "Paper-Launch" AthlonXP 2600 vs Northwood 2.53GHz, days before the release of the Northwood 2.8GHz?
Is it fair to benchmark a "Paper-Launch" AthlonXP 2600 vs Northwood 2.53GHz, days before the release of the Northwood 2.8GHz?
Is it fair for B0pco to share the same address as Intel? Is it fair for B0pco to tweak sysmark 2002 by eliminating all the tasks that the Athlon did well in in Sysmark 2001?
Was it fair to people that they bought P4s with SDRAM while most benchmarks were taken using Rambus?
Is it fair for a class action lawsuit to be filed against Intel for misleading people on performance...heee
Ah, life is just an illusion...
Was it fair to people that they bought P4s with SDRAM while most benchmarks were taken using Rambus?
Is it fair for a class action lawsuit to be filed against Intel for misleading people on performance...heee
Ah, life is just an illusion...
Why would it be absurd? Software that utilizes SSE2 like Lightwave are commonly used. Therefore, it's informative. I bet the people who actually use it and get the performance boost couldn't give a hoot about how it gained its performance. And on another topic, wtf is ScienceMark? Does it have any relevance to what any user would run on his machine?
And btw, even though Intel has not officially embraced DDR333 in their chipset, many of the motherboard manufactuers (including the one Aces used) have gone ahead and tested/validated DDR333 with their boards. So no, it's not running out of spec. It's within perfect spec of the motherboard. The motherboard manufacturer simply felt to go beyond the original reference laid out by Intel.
"Is it fair to compare a paper launched P4 2.8ghz to a paper launched AthlonXP 2600+ ?"
The 2.6, 2.66, and 2.8GHz Northwood CPUs were all available for purchase at Newegg on Monday August 26th, the day of the Anandtech 2.8GHz Review
The 2.6, 2.66, and 2.8GHz Northwood CPUs were all available for purchase at Newegg on Monday August 26th, the day of the Anandtech 2.8GHz Review
Ahh I was wondering when you'd get in on this one Fkloster?
Umm yeah I admit they should make chipsets! If you remeber the AMD 761 was a suberb chipset. They are making chipsets for the hammer, but they should make an updated 761. I hate VIA chipsets for the most part and wish AMD or some other company besides them would take the Athlon chipset market share away from VIA. Nvidia? SIS? Also they desperately need to make a mobile chipset for non intergrated graphics cards.
Also AthlonXP + 512L2 cache + 333fsb = Barton. I think with 512 cache alone AMD will dominate, but the 333fsb just adds to the gap. Also SSE2 optiomisation cannot be ruled either.
All this benchmark flamer stuff changes when someone has to lay their hard earned money down.
How much is a P4-2800? Lot's of paper. XP-2600+ paper too. All paper. EIther you're stupid to pay for a 2800 P4 or can't find a 2600+. No difference.
How much is a P4-2800? Lot's of paper. XP-2600+ paper too. All paper. EIther you're stupid to pay for a 2800 P4 or can't find a 2600+. No difference.
Dude, you like AMD. It's OK. There is no need to justify why AMD is performing worse in many benchmarks.
OKay obviousily some people are missing the point of this post!
First off Newegg always follows paper launches and has processors that they say will be available at a certain date. Well I must conced the point that Intel is much better at getting there processors out to suppliers they are several times larger than AMD.
Secondly DDR333 is not supported "offically" by Intel nor is PC1066 RDRAM. DDR333 is supported in many Athlon chipsets. Many p4 chipsets support DDR333 but lets face it most people buy Intel chipset p4 mobos. SIS is haveing some success but forget about VIA in the p4 market. There are many desktops from mainstream PC companies that have AthlonXP and DDR333. Find me a dell with PC1066 RDRAM? Find me a gateway with DDR333? enough said. Well it is supported on the motherboard???? Isnt the 333fsb supported on Athlon mobos and so are multiplier adjustments? So by your logic it is fair to compare an AthlonXP with 333fsb to an Intel mobo running running at stock fsb? Remember we arent overclocking so we arent running the CPU out of spec!! You can do that with an AMD chip!
First off Newegg always follows paper launches and has processors that they say will be available at a certain date. Well I must conced the point that Intel is much better at getting there processors out to suppliers they are several times larger than AMD.
Secondly DDR333 is not supported "offically" by Intel nor is PC1066 RDRAM. DDR333 is supported in many Athlon chipsets. Many p4 chipsets support DDR333 but lets face it most people buy Intel chipset p4 mobos. SIS is haveing some success but forget about VIA in the p4 market. There are many desktops from mainstream PC companies that have AthlonXP and DDR333. Find me a dell with PC1066 RDRAM? Find me a gateway with DDR333? enough said. Well it is supported on the motherboard???? Isnt the 333fsb supported on Athlon mobos and so are multiplier adjustments? So by your logic it is fair to compare an AthlonXP with 333fsb to an Intel mobo running running at stock fsb? Remember we arent overclocking so we arent running the CPU out of spec!! You can do that with an AMD chip!
Okay I am still pro AMD. AthlonXP only performs worse in benches that arent really cpu specific. Also many of these so called benches are Intel optimized. Lets not forget the Sysmark fiasco! I am bringing to light the issue of testers comparing an RDRAM PC 1066 system which hardly no one will ever have to a more mainstream Athlon setup. A lot of reviewers have made the point of their tests being fair or not to the Athlon.
RE:"Dude, you like AMD. It's OK. There is no need to justify why AMD is performing worse in many benchmarks."
When Intel developed the P4 that actually stated they were going to find benchmraks that made it look good. Now that that has happened people are more than willing to pay extra for less real world performance.
I can undertand the average Joe struggling with the Mhz issue but people here should know better about benchmarks.
When Intel developed the P4 that actually stated they were going to find benchmraks that made it look good. Now that that has happened people are more than willing to pay extra for less real world performance.
I can undertand the average Joe struggling with the Mhz issue but people here should know better about benchmarks.
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 Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter