Yep, and especially good for high refresh rate gaming.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Any Recent Buyers of the 8700K Feeling Buyer's Remorse?
- Thread starter jwilliamson47
- Start date
Yep, and especially good for high refresh rate gaming.
Caporegime
Nah please don't comment too much in areas you are not familiar withYou are not a photographer are you?
Stacking and stitching are commonly used in landscape photography. In this example and this example, it took hours to align, stack and process 64-256 frames with a 4.8GHz Skylake. With Ryzen it would take longer to complete, because the bottleneck would be single-threaded commands.
Photoshop is the de facto industry standard in raster graphics editing. Portrait, landscape, fine art, architecture, commercial, food, watch etc all uses Photoshop.
Use your GPU, its much faster.
Soldato
- Joined
- 7 Dec 2015
- Posts
- 3,043
I have never had a weaker GPU (of the same generation) than you do in your signature, and I have always ticked that "Use Graphics Processor" option by default. You just don't understand which commands/operations cannot utilize the help from the GPU or multi-thread CPU do you?
Caporegime
It works just fine for me, clearly your world consists of Photoshop i get that but photoshop was not what you were talking about when having a dig at Ryzen for Cinebench, its only when you realized that Cinebench is actually a performance checking tool for a real application that sits in a whole world of similar applications that you switched to Photoshop as if it also resides in that world and matters more, it does not. i Use photoshop as one step sometimes in a line of tasks to make textures, there are about 10 other applications as part of those steps and a lot of the time i have half of them doing something all at the same time because you know what? doing one thing at a time is really inefficient.
why would you have remorse at having a faster gaming cpu which will remain faster for the whole period of owning the pc with it in ? seems trolly to me.
Not really, my 8700k has gone into a gaming rig where it is still King. Stronger IPC and ability to hit higher clock speeds works well enough for me in games vs more cores at slower clock speeds. Just kind of frustrating I had to delid to make the most of it vs Ryzen which comes soldiered.
With that said, if I was buying new, would look at the refresh rate I am playing at and if at a lower refresh rate, would take a harder look at Ryzen 2.
With that said, if I was buying new, would look at the refresh rate I am playing at and if at a lower refresh rate, would take a harder look at Ryzen 2.
Soldato
- Joined
- 7 Dec 2015
- Posts
- 3,043
I agree that Ryzen is very powerful for multi-threaded tasks. I also agree that people use Cinebench's multi-thread score to measure that capability. I was just pointing out that Cinebench does provide a single-thread score as well, and that can be used to indicate how well the CPU performs for single-thread-bound tasks.
There are still many algorithms that are difficult to parallelise, such like depth first search, huffman decoding, outer loops of simulated annealing, Dijkstra's shortest path etc. They are either very sequential, or with very low scaling despite high CPU usage with speculative multithreading.
Not a troll post at all, was interested if the same socket, soldered instead of TIM and competitive performance made anyone consider their purchase choice. I own an 8700K and perfectly happy with it.
Associate
Nope, love mine, does exactly what I expect of it.
Caporegime
It exists by request as a benchmarking tool but not actually relevant to it, it never renders on one thread, ever.
It doesn't matter anyway, unless your overclocking, the single threaded IPC is not far behind coffeelake. i don't know how far but in another thread a 6700K just scored about the same in single threaded as the 2700X at the same speed, i'm not even sure there is an IPC uplift from SkyLake to Coffeelake? what does your 8700K score at fixed 4.2Ghz?
Associate
- Joined
- 31 Aug 2017
- Posts
- 2,209
Really doesnt seem to be much in it now, depending on what the cpu is doing of course, but IPC for both sides is now roughly similar.
Higher clocks still puts intel a little higher in some tasks but the new XFR can crank up clocks as well on the 2700x so its all a bit... well kid of similar.
Higher clocks still puts intel a little higher in some tasks but the new XFR can crank up clocks as well on the 2700x so its all a bit... well kid of similar.
Caporegime
IPC is not the same from one workload to another but many people like to use Cinebench as the yardstick, or at least they used to before Ryzen came along.
In MT Ryzen actually scores bar 1% the same as Coffeelake, clock for clock. this is Ryzen 1
In ST Coffeelake was 8% ahead, again this is with Ryzen 1
Ryzen 2 has had an IPC uplift, how much no one knows exactly, perhaps someone with a 2700X and someone with an 8700K can compare notes in Cinebench?
8700K seems to show it's still the best gaming CPU, so I can't see why you'd have remorse about buying it (unless you want more cores for production etc.).
I think I'd feel a little bummed out if I'd paid £240 or so for an 8600K, but that's still a good CPU.
Everyone wins with Ryzen really - Intel and AMD fans - because AMD is producing better products and forcing Intel to accelerate their roadmap.
I think I'd feel a little bummed out if I'd paid £240 or so for an 8600K, but that's still a good CPU.
Everyone wins with Ryzen really - Intel and AMD fans - because AMD is producing better products and forcing Intel to accelerate their roadmap.
Agreed, the competition has helped produce excellent chips on both sides and upped the core count.
Ah right, well I'll have to look closer into multiple ST benchmarks then. I was trying to decide which is better and have seen the AMD CPU's beat Intel in the MT stuff but then I didn't know the 8700K is only a hex core.
Isnt this a duplicate thread? I replied to the same question 2 nights ago.
I own a 8600k and the answer is no.
2 points.
1 - The games I play and my usage pattern, its the best chip for me, I dont play these games which are "unusual" for the PC market, most games I play perform better with maxed out single core performance. The 2 exceptions are FF15 and GTA5. My chip is good enough for GTA5 and not a bottleneck, FF15 is however a game that loves per core performance and more clocks so works better on my chip than chips with only a few more cores, e.g. my 8600k outperforms an 8 core ryzen.
2 - The reviews we seeing have a lot of bench tests done on intel 8 series chips without a user overclock applied, to me this is just weird, people dont usually buy intel K series chips and then just run them at stock, its like buying a ferrari and disabling 2 cylinders in the engine. A key advantage of the intel chips is that they overclock well, so for a reviewer to then just "ignore" that metric is just bizarre. I then think back to what happened on the vega56/64 reviews where reviewers weirdly omitted the 1080ti from graphs, and it came out that was on AMDs request in the review guide, so the only conclusion I have is that AMD requested reviewers to not include overclocked intel chips in the comparison data. These reviewers did test the intel chips overclocked, but they simply did not include the data on their ryzen 2 graphs.
I do think the ryzen 2 chip is now a better choice than 8700 and 8700k if you really value max logical threads, but the i5 is a better chip if you just care about gaming and "normal" desktop use.
However if I owned a 8700k, I wouldnt be needing anti depressants, it is still a better chip than ryzen for most PC games on the market.
I am going to attach a picture of what I mean
So on the pic Imagine red is games that will take advantage of any cores on the system, so called "optimised" games. Blue is games that are usual for the PC gaming market where only 1 or 2 cores get used, because threaded code is difficult to do, lower budget games tend to be in this bracket.
Basically all the reviewed games squeeze into that red area. This is why some of us say the gaming performance review market is skewed.
Also when people say games are capped to 4 cores, thats not actually true, this comment has originated from people with 4 core 8 threaded chips, wondering why a game is not taking advantage of all 8 threads, the answer to that is logical cores dont usually help games. You will find when using a cpu with more than 4 real cores those cores do get utilised in those games. Non threaded games typically either just use one core for "everything", or they will have a rendering thread on one core and a second thread for everything else on another core. About 90% of PC games I play are like this, but about 0% of games performance reviewed are like this.
Also for gaming it really should be the 2600x vs the 8600k, both the 8700k and 2700x are overpriced for minimal benefit over their little siblings.
I own a 8600k and the answer is no.
2 points.
1 - The games I play and my usage pattern, its the best chip for me, I dont play these games which are "unusual" for the PC market, most games I play perform better with maxed out single core performance. The 2 exceptions are FF15 and GTA5. My chip is good enough for GTA5 and not a bottleneck, FF15 is however a game that loves per core performance and more clocks so works better on my chip than chips with only a few more cores, e.g. my 8600k outperforms an 8 core ryzen.
2 - The reviews we seeing have a lot of bench tests done on intel 8 series chips without a user overclock applied, to me this is just weird, people dont usually buy intel K series chips and then just run them at stock, its like buying a ferrari and disabling 2 cylinders in the engine. A key advantage of the intel chips is that they overclock well, so for a reviewer to then just "ignore" that metric is just bizarre. I then think back to what happened on the vega56/64 reviews where reviewers weirdly omitted the 1080ti from graphs, and it came out that was on AMDs request in the review guide, so the only conclusion I have is that AMD requested reviewers to not include overclocked intel chips in the comparison data. These reviewers did test the intel chips overclocked, but they simply did not include the data on their ryzen 2 graphs.
I do think the ryzen 2 chip is now a better choice than 8700 and 8700k if you really value max logical threads, but the i5 is a better chip if you just care about gaming and "normal" desktop use.
However if I owned a 8700k, I wouldnt be needing anti depressants, it is still a better chip than ryzen for most PC games on the market.
I am going to attach a picture of what I mean
So on the pic Imagine red is games that will take advantage of any cores on the system, so called "optimised" games. Blue is games that are usual for the PC gaming market where only 1 or 2 cores get used, because threaded code is difficult to do, lower budget games tend to be in this bracket.
Basically all the reviewed games squeeze into that red area. This is why some of us say the gaming performance review market is skewed.
Also when people say games are capped to 4 cores, thats not actually true, this comment has originated from people with 4 core 8 threaded chips, wondering why a game is not taking advantage of all 8 threads, the answer to that is logical cores dont usually help games. You will find when using a cpu with more than 4 real cores those cores do get utilised in those games. Non threaded games typically either just use one core for "everything", or they will have a rendering thread on one core and a second thread for everything else on another core. About 90% of PC games I play are like this, but about 0% of games performance reviewed are like this.
Also for gaming it really should be the 2600x vs the 8600k, both the 8700k and 2700x are overpriced for minimal benefit over their little siblings.
I was planning to wait and see how the new Ryzens perform but recently Far Cry 5 had the occasional asset streaming spike with all my 7700K threads hitting 100% momentarily and at the same time I found a decent deal for 8700K+motherboard so I went for it. It's a beast of a processor but I'm also really happy to see AMD narrow the gap in gaming (and almost effortlessly bypassing Intel in other areas when comparing processors within similar price brackets).
Having read the reviews if I was to buy a processor today, I'd personally still go for 8700K because high-end gaming is my main purpose for all this stupid hardware and as I found out a quiet (but hefty) air cooler is still enough to overclock this 8700K to 4.8GHz even without delid. I didn't put much effort into this overclock and haven't tried to squeeze more out of it, I'm comfortable with getting the easy MHz out while keeping the core voltage relatively low and the computer quiet. Then again the 2700X actually seems to simply not need overclocking at all, Precision boost/XFR2 do a great job getting the most out of that processor. Really impressive. Digging further into the reviews Ryzen seems to on par in most situations but with some games frametimes seem to still vary more with AMD and there's the odd game that suffers from really low minimum frame rates. A lot better than with the 1800X but the same issue with consistency can still be found even if diminished.
So as for the topic, no regrets. For absolute gaming consistency I made the right choice. For most people out there with need for high-end CPU I'd recommend Ryzen. For price conscious but still want great performance I'd also recommend Ryzen. For my gamer friends it's going to be 8700K for now.
Having read the reviews if I was to buy a processor today, I'd personally still go for 8700K because high-end gaming is my main purpose for all this stupid hardware and as I found out a quiet (but hefty) air cooler is still enough to overclock this 8700K to 4.8GHz even without delid. I didn't put much effort into this overclock and haven't tried to squeeze more out of it, I'm comfortable with getting the easy MHz out while keeping the core voltage relatively low and the computer quiet. Then again the 2700X actually seems to simply not need overclocking at all, Precision boost/XFR2 do a great job getting the most out of that processor. Really impressive. Digging further into the reviews Ryzen seems to on par in most situations but with some games frametimes seem to still vary more with AMD and there's the odd game that suffers from really low minimum frame rates. A lot better than with the 1800X but the same issue with consistency can still be found even if diminished.
So as for the topic, no regrets. For absolute gaming consistency I made the right choice. For most people out there with need for high-end CPU I'd recommend Ryzen. For price conscious but still want great performance I'd also recommend Ryzen. For my gamer friends it's going to be 8700K for now.
just want to point out ryzen 2 is not currently spectre type 2 mitigated, its ok on meltdown, but is in the same boat as intel on spectre. No microcode updates are rolled out yet on ryzen 2 so there is performance hits to come on that if spectre scares you.
Personally I think spectre mitigation is not worth bothering with given its extremely hard to get it to work and the mitigation's offered are only partial.
As far as I know the OS patch for spectre on ryzen 2 doesnt include the microcode unlike intel. But ideally we need someone with a ryzen 2 chip to confirm using inspectre or something.
Personally I think spectre mitigation is not worth bothering with given its extremely hard to get it to work and the mitigation's offered are only partial.
As far as I know the OS patch for spectre on ryzen 2 doesnt include the microcode unlike intel. But ideally we need someone with a ryzen 2 chip to confirm using inspectre or something.