Impressed with FX-8350 and the new article at Anand

[galego]

To engage in a "who can predict the future better" contest is stupid.

Except that the people who is making projections about the near future gaming are the same developers who are making those games.

 

[cbrunny]

Except that the people who is making projections about the near future gaming are the same developers who are making those games.

You missed my point. If anyone can predict the future of gaming it is the people designing the games. My point was specifically referring to this thread and its contributors trying to predict the future.

I'm as hopeful as all other FX8350 owners that this is the future of gaming, but that one source that suggests it is far from concrete. Way too many unknown variables.

Predicting the future is stupid.

 

[galego]

So Intel's middle-grade, two-year-old CPU is ahead of AMD's current top-of-the-line flagship product. Um, did you think this point supported you in some way? :biggrin:

It means that they don't recommend upgrading from a FX-8350 to more expensive i7-3770k because "you may not notice a worthwhile difference in game performance."

It means that the ultra-expensive extreme chips, the expensive i7-3770k and the more cheap and old EoL i5-2300 are all in the same gaming performance tie.

It means that the Haswell i7-4770k will be added to the same tie in a pair of months... but then the old i5-2300 will be still more cheap.

 

[galego]

You missed my point. If anyone can predict the future of gaming it is the people designing the games. My point was specifically referring to this thread and its contributors trying to predict the future.

Ok

 

[jaqie]

It means that they don't recommend upgrading from a FX-8350 to more expensive i7-3770k because "you may not notice a worthwhile difference in game performance."

It means that the ultra-expensive extreme chips, the expensive i7-3770k and the more cheap and old EoL i5-2300 are all in the same gaming performance tie.

It means that the Haswell i7-4770k will be added to the same tie in a pair of months... but then the old i5-2300 will be still more cheap.

No, it doesn't. I just can't see how you can think this, it's confounding how anyone could think this.

 

[frozentundra123456]

It means that they don't recommend upgrading from a FX-8350 to more expensive i7-3770k because "you may not notice a worthwhile difference in game performance."

It means that the ultra-expensive extreme chips, the expensive i7-3770k and the more cheap and old EoL i5-2300 are all in the same gaming performance tie.

It means that the Haswell i7-4770k will be added to the same tie in a pair of months... but then the old i5-2300 will be still more cheap.

The key word is "upgrading". Just because it may not be worth it to upgrade from an 8350 to a 3570 or 3770 is not the same thing as saying the FX is the chip that is best or the one that you should pick if you are starting from scratch.

 

[guskline]

galego, your computer specs please? Why are you avoiding posting them?

 

[galego]

The key word is "upgrading". Just because it may not be worth it to upgrade from an 8350 to a 3570 or 3770 is not the same thing as saying the FX is the chip that is best or the one that you should pick if you are starting from scratch.

The keyword is performance: "you may not notice a worthwhile difference in game performance."

 

[frozentundra123456]

The keyword is performance: "you may not notice a worthwhile difference in game performance."

No, the keyword is "worthwhile" as in worth the cost of a new cpu and motherboard.

The fact that the 8350 is not on their recommended cpu list must mean that they dont think it is the best buy at its price point.

 

[Jaydip]

galego, your computer specs please? Why are you avoiding posting them?

Gusk he is a console gamer and very afraid of Titan :biggrin:

 

[Charles Kozierok]

It means that they don't recommend upgrading from a FX-8350 to more expensive i7-3770k because "you may not notice a worthwhile difference in game performance."

It means that the ultra-expensive extreme chips, the expensive i7-3770k and the more cheap and old EoL i5-2300 are all in the same gaming performance tie.

It means that the Haswell i7-4770k will be added to the same tie in a pair of months... but then the old i5-2300 will be still more cheap.

Well, I just had a short conversation with Don Woligroski about his charts. I even specifically asked him if he really felt the FX-8350 belonged in the same tier as chips like the i3-2120. And he pointed me to this review, which specifically assesses the performance of the FX line against untel in the under-$200 category.

Here's the review's summary chart:

And his conclusion:

For your dollar, the Core i5 has no competition above $160. At $130, the Core i3-3220 is tough to beat. It no longer humiliates the FX line-up in games thanks to AMD's most recent architectural update, but it's still cheaper, faster, and more power-friendly than most of the Vishera-based models.

The FX-8350 beats a two-year-old Deneb chip by less than 1% and loses to the Core i3. Ouch.

His best praise for the AMD chips is:

At least for the time being, whatever quad-core Athlon II and Phenom II processors that are still available seem like smart buys.

You might want to choose a different set of articles for your future propagandizing, because Don's CPU charts just aren't getting the job done for you.

 

[galego]

No, the keyword is "worthwhile" as in worth the cost of a new cpu and motherboard.

No mention to cost but to performance: "you may not notice a worthwhile difference in game performance."

The fact that the 8350 is not on their recommended cpu list must mean that they dont think it is the best buy at its price point.

Congrats for saying the obvious, because nobody here said you that was the best cpu for current games... What is/was being said here is that the 8350 is close to the i7-3770k regarding game performance. One tier in their cpu gaming hierarchy.

And recall this hierarchy is about current games, which are poorly threaded and ignore the true performance of 8-core chips as the 8350.

Crysis 3 is one of a new gen of games that start to use the performance of AMD chips. This is why the FX-8350 is able to outperform the i7-3770k in crysis 3. And as said before all triple-A game developers participating in eurogamer poll selected the FX 8350 as best cpu for future gaming.

 

[toyota]

No mention to cost but to performance: "you may not notice a worthwhile difference in game performance."



Congrats for saying the obvious, because nobody here said you that was the best cpu for current games... What is/was being said here is that the 8350 is close to the i7-3770k regarding game performance. One tier in their cpu gaming hierarchy.

And recall this hierarchy is about current games, which are poorly threaded and ignore the true performance of 8-core chips as the 8350.

Crysis 3 is one of a new gen of games that start to use the performance of AMD chips. This is why the FX-8350 is able to outperform the i7-3770k in crysis 3. And as said before all triple-A game developers participating in eurogamer poll selected the FX 8350 as best cpu for future gaming.

so ONE review shows the 8350 faster in Crysis 3 but you ignore the other reviews that show the 3770k actually being faster in Crysis 3.

 

[guskline]

Good morning galego, Once again, your pc computers specs please? Mine are below both the 3770k and FX 8350. Have you run ANY benchmarks on either machine yourself?

 

[Jaydip]

You are persistent Gusk

 

[Charles Kozierok]

What is/was being said here is that the 8350 is close to the i7-3770k regarding game performance. One tier in their cpu gaming hierarchy.

Wrong. If they were "close" then they would be in the same tier. That's what the tiers mean.

Read the review conducted by Woligroski, the author of the CPU charts you persist in misrepresenting. It is 100% clear that he doesn't consider the FX-8350 a good buy even on a performance per dollar basis.

 

[AtenRa]

Well, I just had a short conversation with Don Woligroski about his charts. I even specifically asked him if he really felt the FX-8350 belonged in the same tier as chips like the i3-2120. And he pointed me to this review, which specifically assesses the performance of the FX line against untel in the under-$200 category.

Here's the review's summary chart:

And his conclusion:



The FX-8350 beats a two-year-old Deneb chip by less than 1% and loses to the Core i3. Ouch.

His best praise for the AMD chips is:



You might want to choose a different set of articles for your future propagandizing, because Don's CPU charts just aren't getting the job done for you.

The review also had Time Latencies which you and Don Woligroski haven’t mentioned. I just wonder why

Raw frame times aren't the end-all in performance analysis because high frame rates have low corresponding frame times and low frame rates have high frame times. What we're trying to look closely at is the lag that exists between consecutive frames in the graphics output.

It really shows how pathetic the Core i3 and avery dual core is for gaming

 

[Charles Kozierok]

The review[/URL] also had Time Latencies which you and Don Woligroski haven’t mentioned.

Yes, I'm sure an experienced, professional tester/reviewer deliberately ignored his own benchmark tests because he's got it in for AMD.

I didn't bring up those CPU charts, galego did. And since he did, the basis for those charts is relevant. The charts are summaries of the individual tests. Averages. Which is what matters. Not your cherry-picking. And he said right in the article that the latencies only really matter when they get above a certain value.

I think I'll take Woligroski's word on this over yours.

 

[Chiropteran]

I think I'll take Woligroski's word on this over yours.

But those charts were Woloigroski's word.



It seems like you accept "his word" when it fits your argument, and reject it when it doesn't. That isn't rational.

 

[AtenRa]

Not your cherry-picking.

Cherry picking ??? those are the charts from his review, i havent cherry picked, i have used all of HIS charts. You on the other hand only used a single chart.

And he said right in the article that the latencies only really matter when they get above a certain value.

And since 90%(or more) of users have 60Hz monitors having more than 60fps doesnt matter, right ??? :whiste:

 

[Charles Kozierok]

Once again, the aggregate chart takes into account the individual tests. So I am taking his word about that chart representing overall performance, over Aten Ra deciding to only focus on a few charts that he thinks make AMD look good.

Frame latencies only matter when they are extremely high. The difference between 1 ms and 2 ms of 95% percentile latency is zero -- it has no impact on actual use.

If you guys have a problem with the review, take it up with Woligroski. Again, I didn't bring this up, I'm just annoyed that galego is deliberately misrepresenting the meaning of his charts.

Anyone here who would rather look at a handful of charts out of a review and then decide on that basis to ignore the summary results and conclusions made by the guy who created those charts is deluding himself.

 

[BrightCandle]

I had an issue when Toms first published a review using this technique and I continue to have the same concern. In the first review they had Intel chips clearly performing better and with more variance, but it was obvious why - they hit higher fps sometimes when the game was CPU limited and when it was GPU limited they achieved the same as the AMD chips. Now obviously by having a higher peak the Intel chips show a higher variance, but its not because there is a low level of microstutter being introduced by the CPU, all that variance came from achieving higher FPS and hence the FPS was less stable. But that is a very different thing to microstutter, that is not something that impacts on any ones game at all.

This review is like the first one I saw with this approach, the measurement of variance while ignoring the fact that the FPS is quite different is highly misleading as to the cause of he variance. The reason of course is that they are measuring the variance wrongly, and it doesn't show what they think it does. Mathematically they don't know what they doing, they have 2 factors both impacting on the number and they need to normalise so only one is being shown.

Its flawed, completely flawed. the reviewer never answered my previous criticism I made against the first review and has carried on, but it doesn't make it any more useful than the first review they did. Its still flawed and the summaries of the stuttering aren't useful.

 

[guskline]

You are persistent Gusk

I'm not trying to be difficult. However, when one posts so much about the comparison of the FX 8350 vs the 3770k and makes so many statements about it, isn't it fair to ask questions such as what are your computer specs and have you ever benched a FX8350 or a 3770k?

AtenRa posts both specs for his FX8350 AND 3770k. He and I may disagree on which one is "better" but I cannot and will not question the fact that he has personal experience with these cpus. That counts a lot as to personal opinion.

Let me say that just because you don't own a particular cpu or gpu doesn't mean you cannot comment on them. However it sure helps my judgment of a posters comments if they have at least had some personal experience with a cpu or cpus they are commenting on.

 

[jaqie]

gus - that's the entire reason I haven't been more active in this thread.

I have experience with the vishera, but not with the 3770k.

I am quite curious to know what his personal system is too, just out of curiosity in my case.

My personal preference is going the way of the dodo - full cores for all. This is why my main system is still a thuban.

 

[Jaydip]

I'm not trying to be difficult. However, when one posts so much about the comparison of the FX 8350 vs the 3770k and makes so many statements about it, isn't it fair to ask questions such as what are your computer specs and have you ever benched a FX8350 or a 3770k?

AtenRa posts both specs for his FX8350 AND 3770k. He and I may disagree on which one is "better" but I cannot and will not question the fact that he has personal experience with these cpus. That counts a lot as to personal opinion.

Let me say that just because you don't own a particular cpu or gpu doesn't mean you cannot comment on them. However it sure helps my judgment of a posters comments if they have at least had some personal experience with a cpu or cpus they are commenting on.

Absolutely Agree :thumbsup: