[Passmark] Ryzen 5 5600X claims the first place in single-thread performance in Passmark

GoodBoy said:
No and I don't measure my penis either...

But thanks for that link, looks like built in goalposts. So point #2 still valid. And why no other results? A bench that leaves out overclocked results seems less useful to me.
Click to expand...
GoodBoy for you perhaps, but we are not comparing overclock vs overclock, comparing stock vs stock here to get baseline measurements. Overclocked results are useless for many because often people can not get the same clocks as the person next to them.
 
d3athf1sh said:
dude, you can't expect them to just give them away... they are now performance leaders. i remember in 2006 the 2.6GHz dual core FX-60 was over $1000
Click to expand...
The issue AMD has is not Intel as much as themselves. Anyone looking to buy an 8 core will likely go after a 3700X, because the performance difference isn't that great from the 5800X. A common theme with AMD is that in a very short amount of time the prices do drop. I'm willing to believe that anybody looking to buy a 6 core CPU for $300 would probably buy the i5-10600K which is currently $270, because they still favor Intel.
 
DukenukemX said:
The issue AMD has is not Intel as much as themselves. Anyone looking to buy an 8 core will likely go after a 3700X, because the performance difference isn't that great from the 5800X. A common theme with AMD is that in a very short amount of time the prices do drop. I'm willing to believe that anybody looking to buy a 6 core CPU for $300 would probably buy the i5-10600K which is currently $270, because they still favor Intel.
Click to expand...
i don't know, with all the security flaws, high power draw/heat, and just all the shady anti-consumer stuff and NO PCIE 4.0. I think i'd rather get a used 3600 and overclock it and upgrade it later to something with more cores or just grab an 8 core from the get if you can afford it.

Funny how all these guys use to joke amd guys before and say i just buy intel because it's the best i don't care who makes it, but now that there's ABSOLUTELY NO REASON to buy an intel platform. The fanboys now start going on the defensive. Hey it's your money but i'm prettty sure you're gonna need pcie 4.0 to run directStorage. So if you're a gamer, one more thing to think about.
 
DukenukemX said:
Anyone looking to buy an 8 core will likely go after a 3700X, because the performance difference isn't that great from the 5800X.
Click to expand...
The used parts segment is only going to serve a small slice of the market. AMD doesn't have a reason to keep making the older parts anymore.
 
Axman said:
The used parts segment is only going to serve a small slice of the market. AMD doesn't have a reason to keep making the older parts anymore.
Click to expand...
When I bought my Ryzen 1700 I paid $200 for a brand new part. My current 2700X I did buy used for $180. As of right now a new 2700X is a little over $200. It's not that much slower than the 3700x either. I do expect the 3700X to be around $250 soon. AMD's stuff drops significantly in prices once new stuff is out.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
d3athf1sh said:
i don't know, with all the security flaws, high power draw/heat, and just all the shady anti-consumer stuff and NO PCIE 4.0. I think i'd rather get a used 3600 and overclock it and upgrade it later to something with more cores or just grab an 8 core from the get if you can afford it.
Click to expand...
The security flaws are bad but for consumers this isn't as much of an issue. I tend to try and disable the security fixes since they can decrease performance significantly because as a home user I don't expect to get hacked like a huge data center. Also, PCIE 4.0? This might be important later on but not right now. AMD's motherboards aren't as future proof either speaking as a B350 owner. By the time you want to upgrade, chances are you can't use the latest AMD CPU, like I can't. I can't use a 3700X and certainly can't use any of the 5000 series Ryzens. I would have to upgrade the motherboard anyway.
 
DukenukemX said:
AMD's stuff drops significantly in prices once new stuff is out.
Click to expand...
And I get that $50, $75, or $100 represents a significant percentage over the previous-gen used parts, especially in the $200 to $300 price range, but if you compare it to other budgets, like food, $20 is fast food for two. $30 to $50 is sit down dinner for two.

I get that times can be tough but that $80 difference also means at least one or two years of future-proofing.

It also makes sense if you need to buy or upgrade right now to buy used, because 1., new parts aren't exactly available, and 2., we're right on the verge of faster hard drives, new sockets, and higher-bandwidth memory. As great a year as 2021 will be for hardware, I also feel like it's going to be a half-step between the best of today's parts as the bog standard for the upcoming gen, and still not as future-proof as DDR5 setups will be, assuming they'll hit the market next year as well.

Which may not be a smart assumption on my part.
 
DukenukemX said:
The security flaws are bad but for consumers this isn't as much of an issue. I tend to try and disable the security fixes since they can decrease performance significantly because as a home user I don't expect to get hacked like a huge data center. Also, PCIE 4.0? This might be important later on but not right now. AMD's motherboards aren't as future proof either speaking as a B350 owner. By the time you want to upgrade, chances are you can't use the latest AMD CPU, like I can't. I can't use a 3700X and certainly can't use any of the 5000 series Ryzens. I would have to upgrade the motherboard anyway.
Click to expand...
Dang that sucks. (About mobo not supporting). A lot of the x370s supported zen 2 (even my cheap x370 asrock)
 
DukenukemX said:
The security flaws are bad but for consumers this isn't as much of an issue. I tend to try and disable the security fixes since they can decrease performance significantly because as a home user I don't expect to get hacked like a huge data center. Also, PCIE 4.0? This might be important later on but not right now. AMD's motherboards aren't as future proof either speaking as a B350 owner. By the time you want to upgrade, chances are you can't use the latest AMD CPU, like I can't. I can't use a 3700X and certainly can't use any of the 5000 series Ryzens. I would have to upgrade the motherboard anyway.
Click to expand...

I mean, that seems to be a limit of YOUR B350 board. The Asus B350-F (which I used to have), supports all the way up to the 3950x, even if it's not necessarily the best idea to drop one in.
 
Axman said:
And I get that $50, $75, or $100 represents a significant percentage over the previous-gen used parts, especially in the $200 to $300 price range, but if you compare it to other budgets, like food, $20 is fast food for two. $30 to $50 is sit down dinner for two.
Click to expand...
There's a reason why sit down diners and restaurants aren't doing well in a pandemic with a recession. Here in NJ there maybe no sit down establishments left after this recession. So yes, $20 for a fast food for two is expensive. You are aware of the amount of people starving right now? Poor choice of an analogy.
I get that times can be tough but that $80 difference also means at least one or two years of future-proofing.
Click to expand...
Not really. Look at the 2500K owners who can still get away with using their CPU's. A 3570K or a 6600K owner is in no better position. The need for a 6600K owner to upgrade is the same for a 2500K, and that's when the need for more cores is needed. So when a Ryzen 1600X owner needs to upgrade will be about the same time a 5600X owner will need to upgrade. As it stands right now a 2500K owner can still avoid upgrading if the clock speed is at 4.4Ghz.
It also makes sense if you need to buy or upgrade right now to buy used, because 1., new parts aren't exactly available, and 2., we're right on the verge of faster hard drives, new sockets, and higher-bandwidth memory. As great a year as 2021 will be for hardware, I also feel like it's going to be a half-step between the best of today's parts as the bog standard for the upcoming gen, and still not as future-proof as DDR5 setups will be, assuming they'll hit the market next year as well.

Which may not be a smart assumption on my part.
Click to expand...
The only reason not to buy used is because some parts can be bad. I've had my fair share of bad hardware I bought used that I didn't know until 1 year later. But I've been burned with new hardware I've bought as well. Things don't evolve as quickly as AMD/Nvidia/Intel would like you to believe. Things like PCIE4 will take years before we'll see any affordable benefits from it.
 
kirbyrj said:
I mean, that seems to be a limit of YOUR B350 board. The Asus B350-F (which I used to have), supports all the way up to the 3950x, even if it's not necessarily the best idea to drop one in.
Click to expand...
Maybe but that still means that AMD doesn't reliably support all of their CPU's for all of their motherboards. Even if I spent extra for a Asus B350-F, I'd still be better off buying a new cheap B550. What's strange is that B550's don't support first gen Ryzen's. That's a limitation of the BIOS size.
 
DukenukemX said:
Maybe but that still means that AMD doesn't reliably support all of their CPU's for all of their motherboards. Even if I spent extra for a Asus B350-F, I'd still be better off buying a new cheap B550. What's strange is that B550's don't support first gen Ryzen's. That's a limitation of the BIOS size.
Click to expand...

AMD technically DOES support the CPUs on various chipsets. Whether or not the board vendors utilize AMD's support is another story. I don't see that as AMD's fault per se. The fact that your board doesn't support the 8 core 3700x seems to be a VRM issue as in the board manufacturer didn't want to chance the higher power draw from Zen2. The 5XX boards definitely have significantly better power delivery (and significantly higher pricing) compared to 1st gen boards.

But yes, there's nothing standing in the way of B550 1st gen support other than the bios size. My X570 will work with a 2400G which is essentially a 1st gen part.
 
VIC-20 said:
Did you even read that graphs you posted? Scroll down. P4 560 held its own against the 3200+ and often the 3400+ as well. The P4 3.4EE still beat a 3500+ in many games.

Athlon 64 socket 754 didn't dominate Intel at all. It traded blows until the very end of their run, with the last couple single core 939 models 3800+/4000+ finally consistently beating Intel in gaming. Then dominating with the X2 and OC'd Opterons

What's with the BS? So many gamers spread the lies of AMD beating Intel in the XP days and then the early 64 days when they were new. And they are still doing it all these years later, despite mountains of evidence that just was not true.
Click to expand...

Lol. That is cute. Asking me if I read the article and then wanting to scroll past athlon64 wins just to cherry pick you precious prescott being able to compete against low/medium end CPU's

Well let us look at the ENTIRE article, shall we?

Here is the benchmark summed up first with who won and how many top positions it had before the competition got a spot

Page 1 Business Winstone 2004 - AMD top 5 positions Communication SYSMark 2004 - AMD top 2 positions Document creations AMD - AMD top 1 postion Data Analysis - Intel top 4 Positions Page 2 Microsoft Office XP SP-2 - None Won Top position Mozilla 1.4 - AMD top 6 positions ACD Systems ACDSee PowerPack 5.0 - AMD top 5 positions Ahead Software Nero Express 6.0.0.3 - AMD top 7 positions Winzip - AMD top 3 positions WinRAR 3.40 - AMD top 6 positions Page 3 MCC Winstone 2004 - AMD top 4 positions 3D Content Creation Sysmark 2004 - Intel top 1 position 2D Content Creation Sysmark 2004 - Intel top 1 positons Web Publication Sysmark 2004 - Intel top 1 positons Mozilla + Media Encoder - AMD top 2 positions Page 4 Adobe Photoshop 7.0.1 - AMD top 4 positions Adobe Premier 6.5 - AMD top 6 positions Roxio VideoWave Movie Creator 1.5 - Intel top 3 positions Page 5 MusicMatch Jukebox 7.10 - Intel top 1 position DiVX 5.2.1 with AutoGK - Intel top 3 positions XViD with AutoGK - AMD top 1 position Windows Media Encoder 9 - AMD top 1 position Page 6 Doom 3 - AMD top 5 positions Counterstrike: Source - AMD top 3 positions Halo - AMD top 3 positions Starwars Battlefront - AMD top 3 positions Battlefield Vietnam - None Won Top position Unreal Tournament 2004 - AMD top 4 positions Page 7 Wolfenstein: Enemy Territory - AMD top 3 positions The Sims 2 - AMD top 5 positions Far Cry - AMD top 4 positions Warcraft III - AMD top 4 positions Page 8 3dsmax 5.1 DX - AMD top 3 positions 3dsmax 5.1 OGL - AMD top 3 positions 3dsmax 6 - rendering composite - Intel top 1 position 3dsmax 6 - 3dsmax5.rays - AMD top 4 positions 3dsmax 6 - Cballs2 - Intel top 2 positions 3dsmax 6 - SinglePipe2 - Intel top 2 positions 3dsmax 6 - UnderWater - Intel top 2 positions Page 9 Visual Studio 6 - AMD top 5 positions SPECviewperf 8 (3dsmax-03) - AMD top 1 position SPECviewperf 8 (catia-01) - AMD top 3 positions SPECviewperf 8 (ensight-01) - AMD top 6 positions SPECviewperf 8 (light-07) - AMD top 3 positions SPECviewperf 8 (maya-01) - AMD top 7 positions SPECviewperf 8 (proe-03) - AMD top 3 positions SPECviewperf 8 (sw-01) - AMD top 3 positions SPECviewperf 8 (ugs-04) - AMD top 3 positions Page 10 Power Consumption IDLE - AMD top 1 position Power Consumption LOAD - AMD top 1 position


OK -you are welcome to go double check before we move on to the cross benchmark analytical part

So lets looks at trading blow vs domination. What I'm going to do here is to hold a scaling threshold between simple winning a benchmark and dominating it based on the top positions. and then you can pick whatever threshold you believe shows a domination position.
2 top wins, 3 top wins, 5 top wins. Doesn't matter YOU get to choose because om so generous ( or because AMD is dominationg so much it doesn't matter)
and then for fun I am also going to look into each page of the article as a singular benchmark and do the same comparison / aka who won the most pages of benchmark. Kinda like an electoral collage for the benchmark because well voting is close

Her we go

Winners.png

P.S. the green/red is not colored for company. It is colored for win/losses. It just fall in line with Intel loosing on all possible comparions because prescott lineup was never "trading blows" with the athlon X64 lineup.


As we can see not only does the athlon64 line up win most benchmark with 3 times as many wins, but also when we increase the threshold for domination. Athlon64 just keep growing its lead. And it should probalby be added that the high number top positons cases... that is not just athlon64 but also Athlon XP beating out the mighty precott 4 EXTREME EDITION. Yeah last generation CPU beating the top of the Line CPU of the competition.

Well as you can also see if I just counted each page for their individual wins (just for fun) then again AMD with its athlon 64 lineup take a strong leadeand the lead keeps growing as we increase the domination treshold

THIS IS WHY WE DONT "SCROLL DOWN AND LOOK AT CHERRY PICKED BENCHMARKS




Now lets looks at this sentence right here
P4 560 held its own against the 3200+ and often the 3400+ as well. The P4 3.4EE still beat a 3500+ in many games.

This is what I hear
In these cherry picked cases the Top of the line CPU could barely beat out a LOW END/ MEDIUM offering. Ergo it is even.
When you have to grasp for anway more expensive extrem editon CPU to barely be holding against the medium of the competitto and consider it trading blows... and then coplete ignored the facts that it also gotten beating by the low end/prior generation cpu... that is what fanboyism does to you.

That is the answer to
What's with the BS?


--- edit ---
As Im going over the benchmark again for a performance percentage win i noticed i accidental gave intel a win in "Microsoft Office XP SP-2" ven thoug they were even. it just happened to be that intel was put first, and i only looked at positions.
so yeah the real number are acutally even worse for Intel than my above cross benchmark analysis

-- edit ---
Make that 2 even benchmark I accidental gave to intel. shits is looking worse for intel as we dig deeper

-- edit ---
I also forgot a benchmark AMD won with 3 positions. so despite butchering up 3 benchmark to Intel favor they still failed horrible...

-- edit ---
and i counted in an intel win twice "3dsmax 6 SinglePipe2 "
so thats 4 benchmark that was butches to intel side and they still came out loosing.

ill see if i can re upload the proper spreadsheet later


--- edit ---
I've adjusted my dta entry for the above metnionedm istakes and redid the spreadsheet
 

Attachments

  • Winners.png
    Winners.png
    6.7 KB · Views: 0
Last edited:
I apologize I was hoping the spoiler tag would hide the long list instead i just blurred it out. but I'm keeping it in my previous post as it is important i think for people to be able to double-check my math in the name of proper testing

Also didn't even cover the fact that I left out power consumption when counting up benchmark. AMD ofcause ruled on that one as we all remember./ which menat if you wanted a lownoise computer intel prescott was NOT the cpu to go for. aka also why it was nicknamed presshot

There is also the fact that i keep in the famous sysmark2004 that we remeber was compiled with ICC so it would remove SSE acceleration on athlon64 CPU.. There is not much point to take another analytical look and remove those as AMD still managed to beat the Intel offerings so horrible hard. but it is still worth mentioning. Please google the AMD vs intel ICC compiler trial and you can read more on this issue
it could be argued the sysmark2004 should Have used the anti intel cheat patch.

Just some quick links on the intel compiler debacle
https://techreport.com/news/8547/does-intels-compiler-cripple-amd-performance/
https://www.osnews.com/story/22683/intel-forced-to-remove-cripple-amd-function-from-compiler/

In short any software compiled with the Intel C Compiler (ICC) would be build with a check to see if the cpu was intel.
If it was Intel it would detect for accelration likes SSE and MMX etc and selecte the correct codepath to run for optimal perfomance
If it was AMD it ignored any accelartion featurs of the CPU and simply only run base X86/x87 operation slowing down software running on AMD CPU's
It took the change of a handful of bytes in the .exe file to bypass this check and regains the full speed of the software on AMD CPU's
 
Last edited:
SvenBent said:
Lol. That is cute. Asking me if I read the article and then wanting to scroll past athlon64 wins just to cherry pick you precious prescott being able to compete against low/medium end CPU's

Well let us look at the ENTIRE article, shall we?

Here is the benchmark summed up first with who won and how many top positions it had before the competition got a spot

Page 1 Business Winstone 2004 - AMD top 5 positions Communication SYSMark 2004 - AMD top 2 positions Document creations AMD - AMD top 1 postion Data Analysis - Intel top 4 Positions Page 2 Microsoft Office XP SP-2 - Intel top 1 positions Mozilla 1.4 - AMD top 6 positions ACD Systems ACDSee PowerPack 5.0 - AMD top 5 positions Ahead Software Nero Express 6.0.0.3 - AMD top 7 positions Winzip - AMD top 3 positions WinRAR 3.40 - AMD top 6 positions Page 3 MCC Winstone 2004 - AMD top 4 positions 3D Content Creation Sysmark 2004 - Intel top 1 position 2D Content Creation Sysmark 2004 - Intel top 1 positons Web Publication Sysmark 2004 - Intel top 1 positons Mozilla + Media Encoder - AMD top 2 positions Page 4 Adobe Photoshop 7.0.1 - AMD top 4 positions Adobe Premier 6.5 - AMD top 6 positions Roxio VideoWave Movie Creator 1.5 - Intel top 3 positions Page 5 MusicMatch Jukebox 7.10 - Intel top 1 position DiVX 5.2.1 with AutoGK - Intel top 3 positions XViD with AutoGK - AMD top 1 position Windows Media Encoder 9 - AMD top 1 position Page 6 Doom 3 - AMD top 5 positions Counterstrike: Source - AMD top 3 positions Halo - AMD top 3 positions Starwars Battlefront - AMD top 3 positions Battlefield Vietnam - Intel top 1 position Unreal Tournament 2004 - AMD top 4 positions Page 7 Wolfenstein: Enemy Territory - AMD top 3 positions The Sims 2 - AMD top 5 positions Far Cry - AMD top 4 positions Warcraft III - AMD top 4 positions Page 8 3dsmax 5.1 - AMD top 3 positions 3dsmax 6 - rendering composite - Intel top 1 position 3dsmax 6 - 3dsmax5.rays - AMD top 4 positions 3dsmax 6 - Cballs2 - Intel top 2 positions 3dsmax 6 - SinglePipe2 - Intel top 2 positions 3dsmax 6 - SinglePipe2 - Intel top 2 positions 3dsmax 6 - UnderWater - Intel top 2 positions Page 9 Visual Studio 6 - AMD top 5 positions SPECviewperf 8 (3dsmax-03) - AMD top 1 position SPECviewperf 8 (catia-01) - AMD top 3 positions SPECviewperf 8 (ensight-01) - AMD top 6 positions SPECviewperf 8 (light-07) - AMD top 3 positions SPECviewperf 8 (maya-01) - AMD top 7 positions SPECviewperf 8 (proe-03) - AMD top 3 positions SPECviewperf 8 (sw-01) - AMD top 3 positions SPECviewperf 8 (ugs-04) - AMD top 3 positions Page 10 Power Consumption IDLE - AMD top 1 position Power Consumption LOAD - AMD top 1 position


OK -you are welcome to go double check before we move on to the cross benchmark analytical part

So lets looks at trading blow vs domination. What I'm going to do here is to hold a scaling threshold between simple winning a benchmark and dominating it based on the top positions. and then you can pick whatever threshold you believe shows a domination position.
2 top wins, 3 top wins, 5 top wins. Doesn't matter YOU get to choose because om so generous ( or because AMD is dominationg so much it doesn't matter)
and then for fun I am also going to look into each page of the article as a singular benchmark and do the same comparison / aka who won the most pages of benchmark. Kinda like an electoral collage for the benchmark because well voting is close

Her we go

View attachment 293140
P.S. the green/red is not colored for company. It is collored for win/losses. It just fall in line with Intel loosing on all possible comparions because prescott lineup was never "trading blows" with the athlon X64 lineup.


As we can see not only does the athlon64 line up win most benchmark with almost 2.5 times as much, but also when we increase the threshold for domination. Athlon64 just keep growing its lead. And it should probalby be added that the high number top positons cases... that is not just athlon64 but also Athlon XP beating out the mighty precott 4 EXTREME EDITION. Yeah last generation CPU beating the top of the Line CPU of the competition.

Well as you can also see if I just counted each page for their individual wins (just for fun) then again AMD with its athlon 64 lineup take a strong leadeand the lead keeps growing as we increase the domination treshold

THIS IS WHY WE DONT "SCROLL DOWN AND LOOK AT CHERRY PICKED BENCHMARKS




Now lets looks at this sentence right here
P4 560 held its own against the 3200+ and often the 3400+ as well. The P4 3.4EE still beat a 3500+ in many games.

This is what I hear
In these cherry picked cases the Top of the line CPU could barely beat out a LOW END/ MEDIUM offering. Ergo it is even.
When you have to grasp for anway more expensive extrem editon CPU to barely be holding against the medium of the competitto and consider it trading blows... and then coplete ignored the facts that it also gotten beating by the low end/prior generation cpu... that is what fanboyism does to you.

That is the answer to
What's with the BS?
Click to expand...
You wrote all that because 1 to 2 FPS difference, LESS THAN 2% until the 3800+ came out much later, IS TRADING BLOWS. WHICH IS WHAT I WROTE.

READING IS FUNDAMENTAL
 
VIC-20 said:
You wrote all that because 1 to 2 FPS difference, LESS THAN 2% until the 3800+ came out much later, IS TRADING BLOWS. WHICH IS WHAT I WROTE.

READING IS FUNDAMENTAL
Click to expand...
I think he wrote all that to explain that an extreme edition top of the line P4 trading blows with the low end of the athlon 64 is not really proving the point that Intel had any kind of advantage at the time.

those of us who were around when the choice was athlon 64 vs p4 were building AMD rigs because they were better processors.

athlon xp vs northwood p4 was a coin toss as to whichever you preferred though.
 
VIC-20 said:
You wrote all that because 1 to 2 FPS difference, LESS THAN 2% until the 3800+ came out much later, IS TRADING BLOWS. WHICH IS WHAT I WROTE.

READING IS FUNDAMENTAL
Click to expand...
You really going to argue that AMD won wit only 1-2% when they have 7 CPU's beating the prescott ? including the athlonXP? or are you again just cherry picking your examples ?
Are you sure you want to make it about percentages ? because i will be happy to go down that road. I love math and despise fanboyisme so its a double win for me if you want to
 
deruberhanyok said:
I think he wrote all that to explain that an extreme edition top of the line P4 trading blows with the low end of the athlon 64 is not really proving the point that Intel had any kind of advantage at the time.

those of us who were around when the choice was athlon 64 vs p4 were building AMD rigs because they were better processors.

athlon xp vs northwood p4 was a coin toss as to whichever you preferred though.
Click to expand...

A P4 560/570 wasn't that expensive (due to the AMD64 competition causing Intel to slash prices) and it wasn't significantly beat by AMD, past a margin of error, until the 3800+ 939s. AMD 64 was impressive in efficiency on release, but not on all out performance. To claim otherwise is pure brand loyalist hyperbole. It was AMD playing catch up, not PWNing anything. Later in 939 that changed. And Intel was flat out destroyed by X2 until Conroe. This is what I said.

I built and sold almost every damn thing that has been released since Pentium II. Thousands and thousands of machines. And I have experience with almost every single release of CPU and chipset. Side by side, P4 vs XP? P4s could always be had faster if you had the cash. P4 vs AMD64? For most of the generation it, they were fairly on par, but that depends on the date you built your PC. At some points, there wasn't an AMD64 that would beat Intel's fastest. At other points that reversed. But it we are talking about under 2% differences. AMD didn't do anything but catch up. Period.

Once 3800+ came out, Intel couldn't keep up anymore. Prescott could not be pushed any further like they planned. Then X2 flat out destroyed them and Pentium Ds went on a fire sale.
 
VIC-20 said:
A P4 560/570 wasn't that expensive (due to the AMD64 competition causing Intel to slash prices) and it wasn't significantly beat by AMD, past a margin of error, until the 3800+ 939s. AMD 64 was impressive in efficiency on release, but not on all out performance. To claim otherwise is pure brand loyalist hyperbole. It was AMD playing catch up, not PWNing anything. Later in 939 that changed. And Intel was flat out destroyed by X2 until Conroe. This is what I said.

I built and sold almost every damn thing that has been released since Pentium II. Thousands and thousands of machines. And I have experience with almost every single release of CPU and chipset. Side by side, P4 vs XP? P4s could always be had faster if you had the cash. P4 vs AMD64? For most of the generation it, they were fairly on par, but that depends on the date you built your PC. At some points, there wasn't an AMD64 that would beat Intel's fastest. At other points that reversed. But it we are talking about under 2% differences. AMD didn't do anything but catch up. Period.

Once 3800+ came out, Intel couldn't keep up anymore. Prescott could not be pushed any further like they planned. Then X2 flat out destroyed them and Pentium Ds went on a fire sale.
Click to expand...

You avoided replying to me if you wanted to check those "past a margin of error,"
What do you consider a margen of error percentage wise , since being the only one bee run over with 6 and 7 competitor options still means "Ttading blows" to you
Seem like deliberately using subjective terms.

You also complete ignored the P4 EE in your response that you yourself brought up.
Lets look at the P4 EE when it having decent performance but lets ignored its price tag and whne it get run over by an old athlon xp3200+
You seems very selective on what data you compare.
 
Hrmmm I have a 3960x or else I would so get a 5800x. Shit I might still build a 5600x and put my 5600xt gpu in it for a low electricity very capable 1080p 240hz rig. And just use the 3960x and 6800xt ( hope to get one ) for high end gaming sessions.
 
VIC-20 said:
A P4 560/570 wasn't that expensive (due to the AMD64 competition causing Intel to slash prices) and it wasn't significantly beat by AMD, past a margin of error, until the 3800+ 939s. AMD 64 was impressive in efficiency on release, but not on all out performance. To claim otherwise is pure brand loyalist hyperbole. It was AMD playing catch up, not PWNing anything. Later in 939 that changed. And Intel was flat out destroyed by X2 until Conroe. This is what I said.

I built and sold almost every damn thing that has been released since Pentium II. Thousands and thousands of machines. And I have experience with almost every single release of CPU and chipset. Side by side, P4 vs XP? P4s could always be had faster if you had the cash. P4 vs AMD64? For most of the generation it, they were fairly on par, but that depends on the date you built your PC. At some points, there wasn't an AMD64 that would beat Intel's fastest. At other points that reversed. But it we are talking about under 2% differences. AMD didn't do anything but catch up. Period.

Once 3800+ came out, Intel couldn't keep up anymore. Prescott could not be pushed any further like they planned. Then X2 flat out destroyed them and Pentium Ds went on a fire sale.
Click to expand...

i think you and I are saying the same thing, then. I’m confused as to how this part of the conversation went where it did.

“if you had the cash”, as you say, was key. If an expensive extreme edition P4 was on par with a notably less expensive low end Athlon 64, or a little faster, and not at all faster than a high end Athlon 64, then I don’t consider that an advantage to Intel.

call it “pure brand loyalist hyperbole” if you want, I guess.
 
Last edited by a moderator:
tangoseal said:
Hrmmm I have a 3960x or else I would so get a 5800x. Shit I might still build a 5600x and put my 5600xt gpu in it for a low electricity very capable 1080p 240hz rig. And just use the 3960x and 6800xt ( hope to get one ) for high end gaming sessions.
Click to expand...
You need a 5600xt cpu to go with your 5600xt gpu. That would satisfy my ocd :)
 
deruberhanyok said:
i think you and I are saying the same thing, then. I’m confused as to how this part of the conversation went where it did.
Click to expand...
One was saying the zen3 release would be like the Athlon XP days, when AMD had an IPC advantage over Intel.

Someone remarked that this time it would be a bit different, ti would not be only an IPC advantage or a performance/budget advantage, unlike the days that when if you were ready to pay a fortune Intel could still keep up or beat AMD in performance, it would be more a like the window when AMD had absolute performance lead regardless on money spent on Intel system could give you.
 
deruberhanyok said:
i think you and I are saying the same thing, then. I’m confused as to how this part of the conversation went where it did.

“if you had the cash”, as you say, was key. If an expensive extreme edition P4 was on par with a notably less expensive low end Athlon 64, or a little faster, and not at all faster than a high end Athlon 64, then I don’t consider that an advantage to Intel.

call it “pure brand loyalist hyperbole” if you want, I guess.
Click to expand...

It really depends on the year/day you pick in time, the AMD board used, chipset used and drivers at the time. Back then those differences really mattered. Your desired mix of gaming/creation useage affected your choice too.

People forget how much P4s would overlock too.

doom3-oc.gif


P4 670s (not shown) could OC fast enough to surpass the even the 4000+ in gaming.


At the time my muddy wanted to build a PC, he wanted the fastest gaming CPU, but no overclocking and money wasn't a concern.

So at moment in time, he bought a Pentium 4 EE.

Here is a review of it beating an FX51 in games (but not synthetic):

https://hothardware.com/reviews/intel-prescott-p4-32ghz-and-p4-ee-34ghz?page=6

Of course, AMD later layed that CPU a big smackdown:

https://www.anandtech.com/show/1517/9

You can find benchmarks of P4 560/570 trading blows with 3200+/3400+. Some review sites showed the P4 faster, sometimes its AMD. Video conversion and workstation benchmarks at the time were often a wider gap. That mattered too. Some review concludes to buy AMD. Some say Intel.

Some they say it depends on your use.

https://www.anandtech.com/show/1517/9


Which is my point. Nobody "owned" nobody back in the 2800+ to 3400+ AMD64 era. Reminds me of AMD 3000 series vs Intel 9000 series. It just depends.

But then came AMD64 X2 and Intel just couldn't keep up at all. No comparison. That is going happen again with the AMD 5000 series CPU releases in my opinion.

Is Intel going to then make a Core 2 like comeback? Maybe. Who knows.
 
VIC-20 said:
It really depends on the year/day you pick in time, the AMD board used, chipset used and drivers at the time. Back then those differences really mattered. Your desired mix of gaming/creation useage affected your choice too.

People forget how much P4s would overlock too.

View attachment 293280

P4 670s (not shown) could OC fast enough to surpass the even the 4000+ in gaming.


At the time my muddy wanted to build a PC, he wanted the fastest gaming CPU, but no overclocking and money wasn't a concern.

So at moment in time, he bought a Pentium 4 EE.

Here is a review of it beating an FX51 in games (but not synthetic):

https://hothardware.com/reviews/intel-prescott-p4-32ghz-and-p4-ee-34ghz?page=6

Of course, AMD later layed that CPU a big smackdown:

https://www.anandtech.com/show/1517/9

You can find benchmarks of P4 560/570 trading blows with 3200+/3400+. Some review sites showed the P4 faster, sometimes its AMD. Video conversion and workstation benchmarks at the time were often a wider gap. That mattered too. Some review concludes to buy AMD. Some say Intel.

Some they say it depends on your use.

https://www.anandtech.com/show/1517/9


Which is my point. Nobody "owned" nobody back in the 2800+ to 3400+ AMD64 era. Reminds me of AMD 3000 series vs Intel 9000 series. It just depends.

But then came AMD64 X2 and Intel just couldn't keep up at all. No comparison. That is going happen again with the AMD 5000 series CPU releases in my opinion.

Is Intel going to then make a Core 2 like comeback? Maybe. Who knows.
Click to expand...
Cool story bro, but no gamer would be caught dead with an EE. Too expensive, no overclock for 90% of them and twice as expensive. Your example does not fit reality for 99% of what real gamers did back then. EE was a super niche product with low sales.
 
Last edited:
Ancient cpu e-peen shouting matches are boring. Hey, I know! Lets discuss Ryzen 5600x/5800x/5950x!

From a TechPowerUp article today:

1603851747366.png


1603851137856.png


The single threaded performance seems quite capable... and check out the RAM that was used to achieve this: DDR4 2400.

So, when does the gaming benchmark embargo lift? Tomorrow?
 
Last edited:
Last edited:
Mega6 said:
.>When does the gaming benchmark embargo lift? Tomorrow?<

LOL - Seriously?
Zen3 November 5
Radeon6K series - to be announced tomorrow
Click to expand...
Well, RX 6000 "Big Navi" launch event is tomorrow... wouldn't be a bad time to show it off running on a 5950x and let slip some benchmarks... no?
 
Mega6 said:
Cool story bro, but no gamer would be caught dead with an EE. Too expensive, no overclock for 90% of them and twice as expensive. Your example does not fit reality for 99% of what real gamers did back then. EE was a super niche product with low sales.
Click to expand...

Jesus this place is toxic.
 
Supercharged_Z06 said:
Well, RX 6000 "Big Navi" launch event is tomorrow... wouldn't be a bad time to show it off running on a 5950x and let slip some benchmarks... no?
Click to expand...
If AMD has the goods, which all the leaks seem to indicate, they will absolutely be showing gaming benchmarks on their best CPU (or maybe the 5600x if gaming performance is good enough there).
 
cybereality said:
If AMD has the goods, which all the leaks seem to indicate, they will absolutely be showing gaming benchmarks on their best CPU (or maybe the 5600x if gaming performance is good enough there).
Click to expand...
I would guess they use the 5800x but we will see.
 
VIC-20 said:
You wrote all that because 1 to 2 FPS difference, LESS THAN 2% until the 3800+ came out much later, IS TRADING BLOWS. WHICH IS WHAT I WROTE.

READING IS FUNDAMENTAL
Click to expand...



Since you are so focused on performance margins but apparently did not want to take me up on the offer and cross benchmark analyzing the performance margins... I did it anyway. just again to show it doesn't matter how you measure it. There is no "trading blows" unless you do the fanboyisme way and cherry pick your benchmarks


First again the data entry:

Business Winstone 200 AMD 13.55% Communication SYSMark 2004 AMD 7.63% Document creation SYSmark 2004 AMD 4.43% Data analysis INTEL 22.60% Microsoft Office XP SP-2 Both 0.00% Mozilla 1.4 AMD 53.75% ACD Systems ACDSee PowerPack 5.0 AMD 15.81% Ahead Software Nero Express 6.0.0.3 AMD 9.98% Winzip AMD 9.98% WinRAR 3.40 AMD 6.75% MCC Winstone 2004 AMD 16.82% 3D content Creation Sysmark 2004 Intel 4.05% 2D content Creation Sysmark 2004 Intel 0.70% Web pulication Intel 2.49% Mozilla + Media Encoder AMD 12.24% Adobe Photoshop 7.0.1 AMD 12.29% Adobe Premier 6.5 AMD 27.35% Roxio VideoWave Movie Creator 1.5 Intel 5.78% MusicMatch Jukebox 7.10 Intel 3.65% DiVX 5.2.1 with AutoGK Intel 11.49% XViD with AutoGK AMD 2.17% Windows Media Encoder 9 AMD 1.16% Doom 3 AMD 16.61% Counterstrike: Source AMD 18.42% Starwars Battlefront AMD 5.71% Battlefield Vietnam Both 0.00% Unreal Tournament 2004 AMD 20.71% Wolfenstein: Enemy Territory AMD 12.09% The Sims 2 AMD 27.46% Far Cry AMD 14.65% Warcraft III AMD 3.33% 3dsmax 5.1 DX AMD 13.56% 3dsmax 5.1 OGL AMD 12.37% 3dsmax 6 Rendering Composite Intel 3.61% 3dsmax 6 3dsmax5.rays AMD 9.28% 3dsmax 6 CBALLS2 Intel 6.76% 3dsmax 6 SinglePipe2 Intel 9.38% 3dsmax 6 UnderWater Intel 4.91% Visual Studio 6 AMD 29.75% SPECviewperf 8 (3dsmax-03) AMD 1.94% SPECviewperf 8 (catia-01) AMD 6.99% SPECviewperf 8 (ensight-01) AMD INF % SPECviewperf 8 (light-07) AMD 7.13% SPECviewperf 8 (maya-01) AMD 51.14% SPECviewperf 8 (proe-03) AMD 9.10% SPECviewperf 8 (sw-01) AMD 10.24% SPECviewperf 8 (ugs-04) AMD 10.32% Power Consumption IDLE AMD 16.27% Power Consumption LOAD AMD 25.44%

Again I'm keeping power consumption out, and is going to show a scaling threshold so you can see it doesn't really matter what you consider an error margin, AMD still comes ahead in the total
Now there is and infinite amount of error margins to choose so a did a handful I found to be good. But since you would probably just come back and say those thresholds was cherry picked. I also went with a no-secrets-up-my-sleeves sequence. which is the Fibonacci sequence


Winners2.png



as we can see it doesn't matter if you believe 1% delta is a win or you need 50% delta is needed to be a win, AMD's Athlon 64 comes out ahead with a staggering ratios starting above 3.0 and as we increase the threshold for a win it just keeps growing reaching up to 14x the win ratio of the Intel offerings before hitting the infinite because the Intel offerings could simple not get a win with as high performance increase as the AMD Athlon 64 could

So, so far we know that :
- AMD Won several times more benchmarks ( close to 3 times as much)
- AMD CPU line has way more CPUs represent in the top of the benchmarks
- When AMD win a benchmark it wins it with a bigger performance leap than Intels P4 lineup ( up to above 50% where Intel will have to settle with below 25%)


Please let me know which way you cross analyzed the benchmark in the article, that you kept telling me to read, and came to the conclusion it was trading blow. And without just cherry picking single benchmarks and ignoring the rest please.

MATH IS FUNDAMENTAL ( yeah I can write big letters too. Am I cool now?)
 
Last edited:
cybereality said:
You're probably wasting your time, the data is clear. People believe whatever they want.
Click to expand...
I am hoping other that looks in will understand how impactive fanboyisme can be on people opinions and always double check facts. It's so annoying when fanboy comes in ruining an otherwise nice debate because they simply dont wont to deal with facts.

Hinting to my post were i was asking for assistance testing the no smt conflicts and no CC jumping tweak to give a boost ot the ryzen cpu, were people like vic-20 would just pot crap about fanboyiem and totally neglect the benchmarks showing the gains I was talking about.
 
You must log in or register to reply here.
Back
Top