HD3870X2 - Intel Q9450 vs Phenom 9850

K

kassler

Limp Gawd
Joined
Jul 10, 2008
Messages
289
Intel Q9450 vs Phenom 9850

Game: Call of Duty 4
Settings: 1920x1200, 4xAA 16xAF 4xAA 16xAF, everything MAX
Q9450 = 65
Phenom 9850 = 69

Game: Crysis
Settings: 1280x1024, 1xAA/1xAF, DX9 everything HIGH
Q9450 = Min FPS 22.46, Max FPS 61.71, Avg 41.96
Phenom 9850 = Min FPS 29.82, Max FPS 63.20, Avg 47.89

Game: STALKER
Settings: 1920x1200, 16xAF everything MAX
Q9450 = 83
Phenom 9850 = 84

Game: TDU
Settings: 1920x1200, 4xAA/16xAF, everything HIGH
Q9450 = 45
Phenom 9850 = 47

Game: Half Life 2
Settings: 1920x1200, 4xAA/16xAF, MAX everything.
Q9450 = 295
Phenom 9850 = 300
 
So q9450 at 2.7 is worse then phenom 9850 for gaming? What gives...
 
I remember something about AMD's chipset doing something with crossfire... I can't remember what though
 
I too like how the AMD chipset performs with AMD/ATI hardware.. @ 3.5ghz/2700nb its just priceless....
 
can we get [H] to do another CPU comparison? especially wiht the new 790GX/FX with SB750 it seems like AMD can overclock their chips quite nicely again...
 
Something's fishy here, I cannot recall ever seeing the Phenoms being more efficient clock-for-clock than core2s, especially 45nm. Any speculation, guys? Does an all-AMD platform really have benefits that great? Or does the 3870X2 just have some strange CPU overhead?

I'm sorry kassler, but I can't read the overclockers club links, don't have flash on my work machine.
 
Waffletron said:
Something's fishy here, I cannot recall ever seeing the Phenoms being more efficient clock-for-clock than core2s, especially 45nm. Any speculation, guys? Does an all-AMD platform really have benefits that great? Or does the 3870X2 just have some strange CPU overhead?

I'm sorry kassler, but I can't read the overclockers club links, don't have flash on my work machine.
Click to expand...

Me neither, does anyone else have proof of this.
 
Waffletron said:
Something's fishy here, I cannot recall ever seeing the Phenoms being more efficient clock-for-clock than core2s, especially 45nm. Any speculation, guys? Does an all-AMD platform really have benefits that great? Or does the 3870X2 just have some strange CPU overhead?
Click to expand...

Here is one explanation why AMD performs better when you run more advanced games and also run them on high res.
It is the same reason as why AMD is strong on servers even if Intel Xeon is higher clocked.
It’s about communication with other hardware (memory and GPU).

When Intel communicates with other hardware it sends data using the FSB. All traffic goes through the FSB. The Latency using the FSB is about 250 clocks. The performance penalty is rather large. If one game is reading memory and/or sends I/O to the GPU using only one thread. Then I don’t think you will see any big differences between AMD and Intel. If the video card is slow you defiantly isn't going see any differences.
Now if the game is using more than one thread the situation changes. If one thread is sending or reading data on Intel, and another thread is sending data to the video card. One of the threads needs to wait. That means that latency goes up. In worst case scenarios it would be double (about 500 clocks).
On AMD this is handled differently. AMD has hypertransport that handles I/O to the video card. If one thread is sending data to the video card it doesn’t compete with memory. On AMD latency is a bit lower too.
Running games on low res or games that isn’t that advanced then this isn’t a problem on Intel. But if you are running a game on high res, have a fast video card (or two). Then this video card is able to handle VERY MUCH I/O. Also if the game is using more memory then more data needs to travel through the FSB on Intel. This situation will lead to more conflicts in the FSB and latency is increased.
Intel is very fast when it isn’t disturbed and is able to use data the cache. More threads, more memory, more communication with video card and maybe more synchronization between threads. Then performance will shift to favor AMD system design.
 
So in real world gaming using quad cores amd is the real winner with phenom cpus?
 
Yes kassler, I know that AMD's design is much better in theory. I just have never seen results that agree with it, only memory benchmarks. It would be interesting to see what happens with a more powerful graphics card in there than a 3870X2. If what kassler is saying is true, the gap should become wider.
 
To test the cpu speed they would need to lower the res some more. All that proved was there was a slight difference in crossfire performance between amd and intel. Both looked gpu bound.

The overclockers club test were interesting though. Very interesting. I would like to see the phenom 9950
 
FlyinBrian said:
To test the cpu speed they would need to lower the res some more. All that proved was there was a slight difference in crossfire performance between amd and intel. Both looked gpu bound.
Click to expand...

Oh! Thats whats going on. Ok, thanks Brian. Its promising to see that AMD can build a good platform. Now we just need nvidia to come back with something to fight. Maybe get actual NATIVE SLI support on intel boards, and then have a platform war! WOO
 
Here we go again. Kassler is the BIGGEST AMD fanboy/troll in existence. Just read his post history.

How about it Kass? Shall we link the countless MORE benchmarks showing Intel > AMD and hear your conspiracy theories again?
 
kassler said:
Here is one explanation why AMD performs better when you run more advanced games and also run them on high res.
It is the same reason as why AMD is strong on servers even if Intel Xeon is higher clocked.
It’s about communication with other hardware (memory and GPU).

When Intel communicates with other hardware it sends data using the FSB. All traffic goes through the FSB. The Latency using the FSB is about 250 clocks. The performance penalty is rather large. If one game is reading memory and/or sends I/O to the GPU using only one thread. Then I don’t think you will see any big differences between AMD and Intel. If the video card is slow you defiantly isn't going see any differences.
Now if the game is using more than one thread the situation changes. If one thread is sending or reading data on Intel, and another thread is sending data to the video card. One of the threads needs to wait. That means that latency goes up. In worst case scenarios it would be double (about 500 clocks).
On AMD this is handled differently. AMD has hypertransport that handles I/O to the video card. If one thread is sending data to the video card it doesn’t compete with memory. On AMD latency is a bit lower too.
Running games on low res or games that isn’t that advanced then this isn’t a problem on Intel. But if you are running a game on high res, have a fast video card (or two). Then this video card is able to handle VERY MUCH I/O. Also if the game is using more memory then more data needs to travel through the FSB on Intel. This situation will lead to more conflicts in the FSB and latency is increased.
Intel is very fast when it isn’t disturbed and is able to use data the cache. More threads, more memory, more communication with video card and maybe more synchronization between threads. Then performance will shift to favor AMD system design.
Click to expand...

Thank you for posting this. You said this better than I ever could. AMD, ftw!
 
Yeah, it's amazing how believable BS sounds when it says what you want to hear. :rolleyes:

If what he's saying is true, that means the 3870X2 (which is not all that powerful compared to what's out now) is already saturating the FSB, which would mean that much more powerful cards like the GTX280 should see very little to no peformance benefit at all on an Intel platform. Why would it if a weaker card is already saturating the FSB?
 
Well, you guys know me. I am not one to defend Kass I want facts and nothing else. The first link was mostly BS but the other links overclockersclub.com shows some interesting results at high resolution giving AMD the lead in some of the games. I am having trouble finding more sources though.
 
FlyinBrian said:
Well, you guys know me. I am not one to defend Kass I want facts and nothing else. The first link was mostly BS but the other links overclockersclub.com shows some interesting results at high resolution giving AMD the lead in some of the games. I am having trouble finding more sources though.
Click to expand...

Yeah, I have to agree with this. AMD may have a more efficient design overall, but their lack of CPU speed/power is what makes them fail against Intel in most situations.

These moments are few and far between, but it is nice to see AMD win every once in a great while. ;)
 
FlyinBrian said:
Well, you guys know me. I am not one to defend Kass I want facts and nothing else. The first link was mostly BS but the other links overclockersclub.com shows some interesting results at high resolution giving AMD the lead in some of the games. I am having trouble finding more sources though.
Click to expand...

Single threaded games will not have any trouble with conflicts in the FSB. There is the latency for each request that is one limiting factor. One game is never 100% bottlenecked by one part, it is many parts that works together. Of course one part could be tha major limiting factor.
But if it takes 100 ns for one computer to send data to another part and 90 ns for another computer. Then the 100 ns computer will be a bit slower than the 90 ns. If the communication is to the DVD and the DVD takes about 10 seconds to respond, then this tiny 10 ns difference will of course not be noticed.

Multithreaded games is another plane. Also how the game is structured. You will se more games coming that scales evenly and that is something that AMD is better at.
 
kassler said:
Single threaded games will not have any trouble with conflicts in the FSB. There is the latency for each request that is one limiting factor. One game is never 100% bottlenecked by one part, it is many parts that works together. Of course one part could be tha major limiting factor.
But if it takes 100 ns for one computer to send data to another part and 90 ns for another computer. Then the 100 ns computer will be a bit slower than the 90 ns. If the communication is to the DVD and the DVD takes about 10 seconds to respond, then this tiny 10 ns difference will of course not be noticed.

Multithreaded games is another plane. Also how the game is structured. You will se more games coming that scales evenly and that is something that AMD is better at.
Click to expand...


Kass, You dont have to over complicate things to make it sound more cool. If AMD do infact offer better high resolution performance. That is awesome, but you dont have to go into a big rant about it and I am trying to say this without being critical and not sounding mean. Really nothing you stated had anything to do with the any possible theory as to why that is.

Now my own personal opinion for what little it is worth is that these games at high resolution are taking advantage of AMD's higher memory bandwith and better I/O. If these test can be confirmed and repeated it would be a good thing for AMD. Right now I havent found many sources on it to verify these findings.
 
I think that FlyinBrian is closer to the truth. The AMD platforms may work better with crossfire, which would make sense. Otherwise these tests would be even MORE dramatic for the 9800GX2, which is a more powerful card than the HD3870X2.

It would be nice if AMD was better at high resolution, but all that I have seen evidence for is that AMD may provide better performace with an ALL AMD platform at high res than a mixed platform. Lets swap in a 9800GX2 and see what happens. Shouldn't high resolution be all GPU limited anyway?
 
Waffletron said:
I think that FlyinBrian is closer to the truth. The AMD platforms may work better with crossfire, which would make sense. Otherwise these tests would be even MORE dramatic for the 9800GX2, which is a more powerful card than the HD3870X2.

It would be nice if AMD was better at high resolution, but all that I have seen evidence for is that AMD may provide better performace with an ALL AMD platform at high res than a mixed platform. Lets swap in a 9800GX2 and see what happens. Shouldn't high resolution be all GPU limited anyway?
Click to expand...

Exactly. Like I said in my previous post, if the all AMD platform was faster for the reasons Kassler is pointing out, adding a more powerful card to an Intel platform would yeild in very little to no performance increase due to the FSB having already been saturated. We know that's not true. The guy is trolling, he has been from day one.
 
wonder how an overclocked Phenom does against the average 9450 @ 3.6

am i reading that link wrong, the phenom doing super PI 1mb in 28 seconds? cus that's TERRIBLE
 
i was just going to mention my grandma's 35$ cpu smokes the 9850 quad core for breakfast w a 18 second superpi 1m time (e2140 @ 3.1ghz)

other than that im sorta impressed if you wanted to run things at stock speeds..
 
Well I don't thinks phenoms super pi time is the point here. What the op was trying to get at is phenom sometimes provides higher fps at high res in multi threaded games.
 
Even if your numbers aren't completely fabricated, all you've done is demonstrated the need for QuickPath.

Here we are four years after the introduction of HyperTransport on the desktop, and the FSB is finally showing signs of saturation. It's a victory that will last all of two months.

What really has me excited is the power savings 45nm seems to offer AMD.
 
You must log in or register to reply here.
Back
Top