- Status
I echo the sentiments of the other commentators but what pleases me more is how a few sites no longer holds the complete voice of the consumers.
It is quite clear form the comment section that individuals are more than capable of comparing and contrasting data to come up with their own conclusions rather than relying on 'expert analysis'.
We all know data can be skewed but I'm just pleased we, as consumers, are able to identify good products for what they are.
Well done Interneters, take a bow.
It is quite clear form the comment section that individuals are more than capable of comparing and contrasting data to come up with their own conclusions rather than relying on 'expert analysis'.
We all know data can be skewed but I'm just pleased we, as consumers, are able to identify good products for what they are.
Well done Interneters, take a bow.
AMD & Intel --> 35nm OK
Intel --> 22nm OK (AMD dropped!)
AMD & Intel --> 14nm OK
Intel --> 10nm seemingly AMD will drop again!)
AMD & Intel --> 7nm in 2020 + PCI Express 4.0 will come
So, what to buy? Actually if you have not an older system of 5-6 years or more, you do not have to update your system. The best computer buying era looks like after 2020 by both AMD and Intel. That time, PCI express will jump to revision 4.0 and I will not update my my desktop and notebook until that time.
Therefore, for now it is not really logical to move to a new system, because new processors only works with Windows 10. In 2020 Microsoft will release Windows 11, will it be fine? I am still using Windows 7, Windows 10 is ridiculus, it looks like ISIS captured my Windows 10 computer because of the black bar, (Their flag is black and white) thus are they captured Microsoft too???
Windows 7 is still much better, being easier to use, but nevertheless, Windows 7 support will end in January 2020. At that time I will think again, to go with Windows or turn to Mac! or what so ever, mobile? Tablet? Anything else?
Intel --> 22nm OK (AMD dropped!)
AMD & Intel --> 14nm OK
Intel --> 10nm seemingly AMD will drop again!)
AMD & Intel --> 7nm in 2020 + PCI Express 4.0 will come
So, what to buy? Actually if you have not an older system of 5-6 years or more, you do not have to update your system. The best computer buying era looks like after 2020 by both AMD and Intel. That time, PCI express will jump to revision 4.0 and I will not update my my desktop and notebook until that time.
Therefore, for now it is not really logical to move to a new system, because new processors only works with Windows 10. In 2020 Microsoft will release Windows 11, will it be fine? I am still using Windows 7, Windows 10 is ridiculus, it looks like ISIS captured my Windows 10 computer because of the black bar, (Their flag is black and white) thus are they captured Microsoft too???
Windows 7 is still much better, being easier to use, but nevertheless, Windows 7 support will end in January 2020. At that time I will think again, to go with Windows or turn to Mac! or what so ever, mobile? Tablet? Anything else?
the nerd 389
Honorable
- Nov 20, 2012
- 1,573
- 5
- 12,460
The differences between the Intel 10 nm and the AMD 7 nm process are going to be fairly murky. The transistor density is slightly better on Intel's process, but the smaller transistors on AMD's should have a very slight efficiency advantage. The efficiency advantage isn't as great as between other process nodes because of the difference in density.
I think the crux of the problem is....you are recommending far worse overall value cpu's in an article that says 'best cpu's'. Even if you make the case that this is for 'gaming only'...it's only an itty bitty bit better on average framerates except for a few older titles...and then only in scenarios with supermega graphics cards that throws the whole value proposition out of the water. You also completely ignore the overclockability of the Ryzen's and the lack of with the Intels and the extra costs of overclocking the 7600K.
Recommending the i5's over the R5's the way you did is doing people a disservice, as they will be much more likely to experience 'gaming' and other limitations much sooner if not immediately.....and the miniscule advantages the i5 might have in very limited scenarios using advanced graphics cards are likely to never materialize.
Recommending the i5's over the R5's the way you did is doing people a disservice, as they will be much more likely to experience 'gaming' and other limitations much sooner if not immediately.....and the miniscule advantages the i5 might have in very limited scenarios using advanced graphics cards are likely to never materialize.
chewy1963
Honorable
- May 9, 2012
- 246
- 0
- 10,680
So Tom's has revealed itself as Intel fanboys. Wonder the exact point point at which Tom's switched from writing for their readers to writing for Intel and other assorted big names (Apple comes to mind)? In other words, when exactly did Tom's jump the shark?
the nerd 389
Honorable
- Nov 20, 2012
- 1,573
- 5
- 12,460
Two conventions, at least one major product release, and anything else that needed to be reviewed?
Well, the title should read "Best Consumer or Mainstream CPUs for 2017." There are Enthusiast consumers out there who are far more tech savvy who visit this site often, and to hear that THE best is an i7 7700k really ignores them and the more powerful cpus they might consider better. I wouldn't have any problem with the article if what is considered best refers to the average consumer.
the nerd 389
Honorable
- Nov 20, 2012
- 1,573
- 5
- 12,460
I think the best way to sum up the popular opinion of this article is:
While the list is technically correct for gaming, gaming is no longer the only metric that enthusiasts use to make decisions. If the other metrics that enthusiasts use are ignored, as they are in this article, it seriously compromises the utility of the resulting advice.
In the past, gaming alone was sufficient to have solid recommendations. This is no longer the case. The approach used to compile this list is obsolete, and in desperate need of an overhaul.
Because this list uses an outdated approach, it does not live up to the expectations that readers of TH have of reviews on this site. If Tom's Hardware wishes to retain it's long-held standing as a review site of the highest caliber, it should consider new approaches for compiling this list.
In many other fields, Tom's Hardware has been a pioneer of new and relevant metrics. To have this article positioned as a key reference in it's current state works against the core ideology of the site as a whole. Continuing to defend the methodology of this article only amplifies the damage to Tom's Hardware's reputation.
Recent developments in the CPU space make the need for a comprehensive and relevant metric ever more necessary. If you cannot make solid and generally applicable recommendations with your current approach, find a new approach.
While the list is technically correct for gaming, gaming is no longer the only metric that enthusiasts use to make decisions. If the other metrics that enthusiasts use are ignored, as they are in this article, it seriously compromises the utility of the resulting advice.
In the past, gaming alone was sufficient to have solid recommendations. This is no longer the case. The approach used to compile this list is obsolete, and in desperate need of an overhaul.
Because this list uses an outdated approach, it does not live up to the expectations that readers of TH have of reviews on this site. If Tom's Hardware wishes to retain it's long-held standing as a review site of the highest caliber, it should consider new approaches for compiling this list.
In many other fields, Tom's Hardware has been a pioneer of new and relevant metrics. To have this article positioned as a key reference in it's current state works against the core ideology of the site as a whole. Continuing to defend the methodology of this article only amplifies the damage to Tom's Hardware's reputation.
Recent developments in the CPU space make the need for a comprehensive and relevant metric ever more necessary. If you cannot make solid and generally applicable recommendations with your current approach, find a new approach.
I don't even think this list of cpu's is correct from a 'gaming' perspective. Even exclusively 'gamers' are getting a worse value with these picks. Gamers will be running into more limitations with the i5's more than they will with the R5's.
As a gross analogy....would it make sense to recommend to gamers a cpu that beats the competition by 1% on specific current and older games while losing by 300% in everything else and almost certain to lose in most newer games. While this is a complete exaggeration on my part, I want to illustrate the point that the advantages the i5's have in games is small enough and questionable enough given the variables that to recommend them over the R5's even to 'gamers' is doing 'gamers' a disservice.
As a gross analogy....would it make sense to recommend to gamers a cpu that beats the competition by 1% on specific current and older games while losing by 300% in everything else and almost certain to lose in most newer games. While this is a complete exaggeration on my part, I want to illustrate the point that the advantages the i5's have in games is small enough and questionable enough given the variables that to recommend them over the R5's even to 'gamers' is doing 'gamers' a disservice.
the nerd 389
Honorable
- Nov 20, 2012
- 1,573
- 5
- 12,460
Ah, the McNamara fallacy. It's easy to ignore variables that aren't quantified.
What's not easy is to actually create an approach that avoids the fallacy. That said, Tom's Hardware has a better shot than most. Just look at their contributions to gaming experience analysis. (Frame time variance, for instance)
What's not easy is to actually create an approach that avoids the fallacy. That said, Tom's Hardware has a better shot than most. Just look at their contributions to gaming experience analysis. (Frame time variance, for instance)
the nerd 389
Honorable
- Nov 20, 2012
- 1,573
- 5
- 12,460
I stand corrected. However, adopting a well designed metric within a week of it's conception is generally considered to be on the leading edge.
Tom's Hardware has been a premier source of tech information since it was founded in 1996. 20 years later, it's still one of the best resources around. Just look at the coverage of the 7900x and compare it to other tech reviewers. Tom's is one of the only sites that actually discusses the thermal flaws of that model in any depth. Other sites bury it in the text or simply say that they're using a new cooling solution with it and not saying why.
Tom's still stands among the few sites unafraid to call manufacturers on their BS (at least in the full reviews). All I'm saying is that this article stands in stark contrast to that tradition.
Tom's Hardware has been a premier source of tech information since it was founded in 1996. 20 years later, it's still one of the best resources around. Just look at the coverage of the 7900x and compare it to other tech reviewers. Tom's is one of the only sites that actually discusses the thermal flaws of that model in any depth. Other sites bury it in the text or simply say that they're using a new cooling solution with it and not saying why.
Tom's still stands among the few sites unafraid to call manufacturers on their BS (at least in the full reviews). All I'm saying is that this article stands in stark contrast to that tradition.
No "Mcnamara fallacy" here... The Nerd 389. The i5 measurably beats the R5 in such a small way when looking at a broad set of games as to be statistically meaningless when recommending a cpu that is completely tapped out compared to an R5. If anything, the R5 might come out ahead using current games and avoiding statistical outliers like Rise of the Tomb Raider...that's at stock.
The R5's win when overclocked, can be overclocked ..unlike most of the i5's recommended...and they convincingly and measurably win at multithreaded tasks at stock. The R5 has loads of reserve power relative to the i5.
Simply put...they are close enough that if you just pick a few of the right games...either could win, but if you look at all the reviews out there, and you consider the pro's and con's of each and the socket and overclockability and upgradeability, if you had a loved one that wanted a processor to just 'game'...you would recommend the R5's.
Tom's dropped the ball and made bad recommendations for gamers here....it's that simple.
The R5's win when overclocked, can be overclocked ..unlike most of the i5's recommended...and they convincingly and measurably win at multithreaded tasks at stock. The R5 has loads of reserve power relative to the i5.
Simply put...they are close enough that if you just pick a few of the right games...either could win, but if you look at all the reviews out there, and you consider the pro's and con's of each and the socket and overclockability and upgradeability, if you had a loved one that wanted a processor to just 'game'...you would recommend the R5's.
Tom's dropped the ball and made bad recommendations for gamers here....it's that simple.
- Status
TRENDING THREADS
-
-
-
-
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
Question No POST, BIOS and video output after SSD. Reinstalling HDD makes everything work.
- Started by Fra-17x
- Replies: 2
Facebook
Twitter
Instagram