Anand: PII vs. Q9550 vs. i7 crossfire, Phenom II = smoother

[Mav451]

Let me ask you rabid Intel supporters posting here, why do you care so much? What if the Phenom platform was actually able to deliver a more enjoyable gaming experience? Would that make you consider it? Are you even open to that possibility? And if you did see graphs that illustrated this, would you actually believe them?

I've been AMD only since I started building, 2001 (AMD T-bird). I patiently waited until Fall 2007 for AMD to reveal it's "Core2 killer". Yeah that went well.

I think 6 out 7 years of my computing experience being pure AMD demonstrates that I'm not a blinded Intel fan, simply a very disappointed AMD fan (and basically have been since summer 2006).

[Zucker2k]

A better way to put it would be, remove the guess work and try both platforms, look at the data, then decide for yourself. That is what matters.

Pounding people over the head and insisting that the raw numbers are the only thing that matters is silly. It also depends on what you are doing. For example, I have an Intel quad box, for encoding work it gets the job done quite a bit faster than my AMD system. But for gaming, I prefer the AMD box, games "feel" smoother, more enjoyable, which actually surprised me. I was certain it would be the opposite.

Let me ask you rabid Intel supporters posting here, why do you care so much? What if the Phenom platform was actually able to deliver a more enjoyable gaming experience? Would that make you consider it? Are you even open to that possibility? And if you did see graphs that illustrated this, would you actually believe them?

I think the subjectivity may lie in the software (games); which brings us to square one.

[roofsniper]

I've been AMD only since I started building, 2001 (AMD T-bird). I patiently waited until Fall 2007 for AMD to reveal it's "Core2 killer". Yeah that went well.

I think 6 out 7 years of my computing experience being pure AMD demonstrates that I'm not a blinded Intel fan, simply a very disappointed AMD fan (and basically have been since summer 2006).

if you are disappointed in phenom II then i really don't know what to say.

[Mav451]

if you are disappointed in phenom II then i really don't know what to say.

Now now, I didn't say that. I said I was disappointed in 2007 (Phenom I). Who knows what this year will bring

(and in particular what this study could reveal)

[eleeter]

I think 6 out 7 years of my computing experience being pure AMD demonstrates that I'm not a blinded Intel fan, simply a very disappointed AMD fan (and basically have been since summer 2006).

I know how you feel. I built my Intel Quad and felt pissed off at AMD every second I was assembling it. But I ended up building a Phenom box out of curiosity mostly. What I didn't expect is using the identical video card, I preferred gaming on the AMD. Keep in mind I only have an overclocked Q6600, but the AMD chip is supposed to be a complete turd in gaming, if you only go by the benchmarks.

I would be curious to hear from someone that has both a Phenom 2 and an i7 that plays some of the latest games. If the i7 gives the nice, smooth gaming as the Phenom I would probably use the i7 as my next build, considering in some areas it pummels the AMD chip.

[Zucker2k]

I know how you feel. I built my Intel Quad and felt pissed off at AMD every second I was assembling it. But I ended up building a Phenom box out of curiosity mostly. What I didn't expect is using the identical video card, I preferred gaming on the AMD. Keep in mind I only have an overclocked Q6600, but the AMD chip is supposed to be a complete turd in gaming, if you only go by the benchmarks.

I would be curious to hear from someone that has both a Phenom 2 and an i7 that plays some of the latest games. If the i7 gives the nice, smooth gaming as the Phenom I would probably use the i7 as my next build, considering in some areas it pummels the AMD chip.

What were your respective overclocks on the Phenom1 and the Q6600?

[roofsniper]

What were your respective overclocks on the Phenom1 and the Q6600?

his phenom is in his sig.

[eleeter]

What were your respective overclocks on the Phenom1 and the Q6600?

q6600@~3.3

[YukonTrooper]

I've been AMD only since I started building, 2001 (AMD T-bird). I patiently waited until Fall 2007 for AMD to reveal it's "Core2 killer". Yeah that went well.

I think 6 out 7 years of my computing experience being pure AMD demonstrates that I'm not a blinded Intel fan, simply a very disappointed AMD fan (and basically have been since summer 2006).

Same here. Phenom left a real bad taste in the mouths of many of AMD's fans, myself included. The performance lacking is one thing, but all the hoopla that led up to the eventual fiasco really poured salt into the wound. It backed many of us into a corner, and the chip became accessible only to the real loyal fans and those not in the know. Phenom II has put AMD back on the map, and I would have purchased such a system if it came two years earlier, but that just didn't happen.

Anyways, I'm an AMD fan simply out of being a fan for the smaller company, so not necessarily an AMD fan, if that makes any sense.

Maybe AMD should market smoothness on their next line of chips. Give it a name and brand it as some kind of technology. "Get the most out of your games with SmoothFrag technology!" Dun, dun, dunnnnn.

[Barrok]

I am in the same boat. I was a bigtime AMD Fan, in fact I purchased a fair amount of their stock a few years ago. Then the stock collapsed, Core2Duo/quad was too good to pass up and now with i7 leading the pack I couldn't go from a q6600 to a p2, even though the p2 was probably better. Felt more like a side step than a step forward

[Sparky]

Same here. Phenom left a real bad taste in the mouths of many of AMD's fans, myself included.

same... to add insult to injury for me was the 9600be I got was a defective chip too, very frustrating. That's why I got the q6600. Now I'm itching to go back since while this q6600 works it sometimes has been a little grumpy with me, don't know why. Plus, it is probably one of the worst clocking q6600 chips out there

[FoxRacR17]

Same here. Phenom left a real bad taste in the mouths of many of AMD's fans, myself included. The performance lacking is one thing, but all the hoopla that led up to the eventual fiasco really poured salt into the wound. It backed many of us into a corner, and the chip became accessible only to the real loyal fans and those not in the know. Phenom II has put AMD back on the map, and I would have purchased such a system if it came two years earlier, but that just didn't happen.

Anyways, I'm an AMD fan simply out of being a fan for the smaller company, so not necessarily an AMD fan, if that makes any sense.

Maybe AMD should market smoothness on their next line of chips. Give it a name and brand it as some kind of technology. "Get the most out of your games with SmoothFrag technology!" Dun, dun, dunnnnn.

I'm upgrading from my good old S939 3800+ X2 OCed to 2.5GHz for about 4 years I think on an Abit KN8-SLi. This was the first computer I ever built and it made me an AMD fan. I ALMOST went i7 this time around, but it just didnt feel right going intel, I felt like I needed to help out AMD, cause if they cave like all businesses seem to be doing in this economy, then I hate to see what kinda prices intel will inflate their chips to. I also like to cheer for the underdog as well. I will have all my parts for my PII build by tomorrow and hopefully have it together next week. I really hope I will not be disappointed.

[marten_larsson]

Now you're really going off topic here - just informing you!

I think the whole data collecting should be done on different systems/setups in different games. This way the phenomenon should be more "real life validated". Sure, a lot of issues arise when doing something like that. One could easily claim that one system had privileges but that's how it is anyway. Of course setups with similar specs should also be included but I think this is such a large "issue" that it needs to be shown with several setups and that some of those are obviously "inferior" to others but still performing better (as in smoother).

Anandtech stated that the smoothness was better on PhII in some games, not all, and therefore more than one game should be tested.

[akaBruno]

Nothing off-topic about preferring a smooth ride in the economy over a jerky computer outfit.

[TL1000S]

This is not a thesis. Do you think you could write your thesis on automobiles, I could read it, and find out what the best car out there was? You probably do.

Hehe... that was actually a good argument.

[TL1000S]

I think the subjectivity may lie in the software (games); which brings us to square one.

I am pretty sure this issue about smoothness boils down to software(drivers).
A couple of us have *tried* to bring this into this pointless "namecalling-thread".
Another example from me:
As I wrote in another post I swapped my DFI board with an Asus one, paired with my new X4 940.
It "stuttered" terrible in 3D-benches, but the score was OK (3DM01).
Then I (re)installed the AMD CPU driver (no such sw exist per se for Intel), voila,.. no stuttering.. just smooothness..
And my 3DM01-score (same clock) went down 10%.
(No effect on 3DM06-score).

[Hornet331]

I am pretty sure this issue about smoothness boils down to software(drivers).
A couple of us have *tried* to bring this into this pointless "namecalling-thread".
Another example from me:
As I wrote in another post I swapped my DFI board with an Asus one, paired with my new X4 940.
It "stuttered" terrible in 3D-benches, but the score was OK (3DM01).
Then I (re)installed the AMD CPU driver (no such sw exist per se for Intel), voila,.. no stuttering.. just smooothness..
And my 3DM01-score (same clock) went down 10%.
(No effect on 3DM06-score).

When it really boils down to drivers, the whole discussion would be again a non issue (yeah keithlm i said it again).
Cause if it is only caused by software and has nothing to do with the hardware itself, the data that might be produced in this thread is a nice bonus, but won't show anything.
Especial when both test systems are configured probably and have a clean install off the OS and the newst drivers.

[gosh]

When it really boils down to drivers, the whole discussion would be again a non issue (yeah keithlm i said it again).
Cause if it is only caused by software and has nothing to do with the hardware itself, the data that might be produced in this thread is a nice bonus, but won't show anything.
Especial when both test systems are configured probably and have a clean install off the OS and the newst drivers.

Software informs the hardware how the hardware should work, software needs to adapt how the hardware likes to work, if software informs the hardware in a way that isn't effective it will not work good.
A graphics driver packs GPU data before it sends it to the gpu. Sending data to the video card is a high latency operation and if you send many small packets it will be much slower compared to sending one big packet. This is much more important on Core 2 compared to Phenom because the FSB likes long trains of data. Core 2 only have one way where data is sent or read from other hardware too. If the software wasn't sending data in long trains (leaving gaps for other data to travel) using the FSB but instead splits the data in many small packets it would be able to the same amount of data. When the FSB is moving data in different directions it also slows down.
If it also need to move memory the traffic increases.
Mouse data is also sent using the FSB on Core 2.
It could be that Core 2 isn't that exact compared to Phenom because Phenom has more capacity moving data to and from the CPU, I/O data doesn't compete with memory traffic.
When I have done heavy work on Intel this is the problem, the speed is often ok. Things get done as fast or maybe faster compared to Phenom but it is trouble to work with other applications when some applications is working hard and the first that starts to behave strange is the mouse. If I start one application that stresses the CPU only it isn't a problem. Thread switching works ok and no problem with data sent to and from the CPU. But if the traffic to and from the CPU increases the mouse will not move as it used to do.

I think that games uses the DirectX driver to read mouse data (don't think they use WIN32 events or API's, haven't read about it so don't know), it also uses the driver to send data to the GPU. It could be that if one have done this badly for that type of hardware the result is that the game will not work well.

[Titan7171]

I dont think it has anythng to do with software, if thats the case you saying that AMD just released this cpu and has the drivers and software running like butter but Intel cant get their hardware to run smooth with all the time theyve had?.........Im not buying it.

[Scubar]

Maybe AMD should market smoothness on their next line of chips. Give it a name and brand it as some kind of technology. "Get the most out of your games with SmoothFrag technology!" Dun, dun, dunnnnn.

I dont think that would work. Intels marketing is far more effective as they have all the facts and figures to backup their claims that they have the top chips on the market right now. You cant really try to counter a big company like Intel by using fictional marketing techniques.

AMD didnt even seem to try and do the same when they had total dominance over Intel back in the Athlon days which is back when i was on the green team for 4-5 years.

[gosh]

Maybe AMD should market smoothness on their next line of chips. Give it a name and brand it as some kind of technology. "Get the most out of your games with SmoothFrag technology!" Dun, dun, dunnnnn.

It is a message that is too complex to market because average user don't have enough knowledge to understand it. Marketing needs to be simple.

[gOJDO]

The toppic although useless, is missleading.
Using 1 graphics card, 4870.

Phenom II 940 isn't "smoother" than i7 920. If you define "smoothness" as min framerate, then P2 940 MIN framerate is 3% slower than I7 920 and 5% higher then Q9550. If you look at the average framerates, the P2 940 is 2% slower in average of all titles than the i7 920, and 1% slower than the Q9550.

[TL1000S]

I dont think it has anythng to do with software, if thats the case you saying that AMD just released this cpu and has the drivers and software running like butter but Intel cant get their hardware to run smooth with all the time theyve had?.........Im not buying it.

You don't have to "buy it".

Then again, I'll give you a simple question:

What can a CPU (Intel or AMD) do without a chipset, RAM, "firmware" (BIos) and an operating system (drivers/programs)?

Just like the "smoothness" issue and other subjective experience, I only tell you what I just can see with my own eyes (and have done before).
Certain 3D-apps do not work "smooth" without the AMD specific drivers (even though I never use Q&Q when benching).
Strange.. very strange...

[Hornet331]

A graphics driver packs GPU data before it sends it to the gpu. Sending data to the video card is a high latency operation and if you send many small packets it will be much slower compared to sending one big packet. This is much more important on Core 2 compared to Phenom because the FSB likes long trains of data. Core 2 only have one way where data is sent or read from other hardware too. If the software wasn't sending data in long trains (leaving gaps for other data to travel) using the FSB but instead splits the data in many small packets it would be able to the same amount of data. When the FSB is moving data in different directions it also slows down.
If it also need to move memory the traffic increases.
Mouse data is also sent using the FSB on Core 2.

I think you forget about that thread already....

It could be that Core 2 isn't that exact compared to Phenom because Phenom has more capacity moving data to and from the CPU, I/O data doesn't compete with memory traffic.
When I have done heavy work on Intel this is the problem, the speed is often ok. Things get done as fast or maybe faster compared to Phenom but it is trouble to work with other applications when some applications is working hard and the first that starts to behave strange is the mouse. If I start one application that stresses the CPU only it isn't a problem. Thread switching works ok and no problem with data sent to and from the CPU. But if the traffic to and from the CPU increases the mouse will not move as it used to do.

There are hardly any desktop apps that can saturate the FSB fully with data, its another topic for high bandwidth server oriented tasks, but thats not what games do. Jack already showed you that theres next to no difference between a FSB of 200mhz and 333mhz and HT/IMC on P1.

I think that games uses the DirectX driver to read mouse data (don't think they use WIN32 events or API's, haven't read about it so don't know), it also uses the driver to send data to the GPU. It could be that if one have done this badly for that type of hardware the result is that the game will not work well.

Uhm.. no, the mouse dont uses direc x as input api, since xp they use the usb stack for HIDclass devices (or the bluetooth stack if its a wireless mouse) and therefor they use the HID.dll as api,
for none HIDclass devices (PS/2) they us a extra drivers.

Direct X or better what you mean, DirectInput, is for Joysticks and ForceFeedback controllers.

I think you where joking about that, cause it took me 3min to find what api/driver is used for mouse input and im not a programmer....

[BenchZowner]

Don't bother replying to the troll man.