Play with anything but top tier GPUs and it won't matter.Assassins Creed Origins needs to be in more CPU gaming benchmarks as that game is a CPU hog. It makes pretty heavy use of all of my 6 cores on the 8700k and from what I had read it gets 4 core cpus up to 100% at times. That game must destroy old i5 cpu's.
the 7600K played Assassin's Creed Origins just fine and the section of the game we tested saw zero frame hitches. We’re keen to explore the game more, but for now we are satisfied with the testing done.
Assassins Creed Origins needs to be in more CPU gaming benchmarks as that game is a CPU hog. It makes pretty heavy use of all of my 6 cores on the 8700k and from what I had read it gets 4 core cpus up to 100% at times. That game must destroy old i5 cpu's.
Play with anything but top tier GPUs and it won't matter.
Assassin's Creed Origins: How Heavy Is It on Your CPU?Assassins Creed Origins needs to be in more CPU gaming benchmarks as that game is a CPU hog. It makes pretty heavy use of all of my 6 cores on the 8700k and from what I had read it gets 4 core cpus up to 100% at times. That game must destroy old i5 cpu's.
Anyway, getting back on track, the 7600K played Assassin's Creed Origins just fine and the section of the game we tested saw zero frame hitches. We’re keen to explore the game more, but for now we are satisfied with the testing done.
It was disappointing to find that the Ryzen 7 1800X and R5 1600X couldn't hold a candle to the Core i5-8400 in spite of the rumors circulating the Internet. As many of you know we really like the Ryzen 5 1600 as a balanced/best value offering here at TechSpot and while it did lay waste to the 7600K, Intel’s rapidly aging quad-core still provided playable performance. Meanwhile the HT enabled Core i7-7700K was able to keep the 1600X in its rear view mirrors.
Benchmarks are intended PURPOSELY not to make initial intuitive sense. That way you can make an actual comparison that matters for future use cases when future games make the 1080Ti whimper and cry. Otherwise, that is just a GPU test and forget about CPU selection.Not sure if some of the posted benchmarks make much sense. I would assume that anyone looking to save money by purchasing a ~$200 CPU wouldn't be in the market for a 1080Ti. Those that are in the market for a 1080Ti probably wouldn't be gaming on a 1080p monitor. I'm not saying those scenarios wouldn't happen but I'll bet they're a small minority.
That said, the 8400 does represent a very good bang-for-your-buck gaming CPU.
Not sure if some of the posted benchmarks make much sense. I would assume that anyone looking to save money by purchasing a ~$200 CPU wouldn't be in the market for a 1080Ti. Those that are in the market for a 1080Ti probably wouldn't be gaming on a 1080p monitor. I'm not saying those scenarios wouldn't happen but I'll bet they're a small minority.
That said, the 8400 does represent a very good bang-for-your-buck gaming CPU.
Benchmarks are intended PURPOSELY not to make initial intuitive sense. That way you can make an actual comparison that matters for future use cases when future games make the 1080Ti whimper and cry. Otherwise, that is just a GPU test and forget about CPU selection.
If you want full system tests (CPU + GPU both in the same price category), then there are plenty of those. But you gain no information about a CPU that way.
It shows the potential of the CPU with the GPU bottleneck removed.
Sure if you run a 1050Ti or 1060 you'll basically see a flat line across all currently available CPUs - that isn't informative at all.
One thing to keep in mind is that CPUs normally last a few GPU upgrade cycles in gaming systems.
It's always better to have some headroom to better utilise future GPUs.
With Volta coming in 2018 we aren't that far away from 1080 Ti level of performance coming down to more mainstream price points.
The plain numbers might be very confusing and might not tell you much,but looking at a video like this how can there still be questions?Not sure if some of the posted benchmarks make much sense. I would assume that anyone looking to save money by purchasing a ~$200 CPU wouldn't be in the market for a 1080Ti. Those that are in the market for a 1080Ti probably wouldn't be gaming on a 1080p monitor. I'm not saying those scenarios wouldn't happen but I'll bet they're a small minority.
That said, the 8400 does represent a very good bang-for-your-buck gaming CPU.
I don't see how this post is relevant to what Elfear was saying but looking at a screenshot with performance metrics definitely doesn't answer the question whether a CPU would be future-proof or not.The plain numbers might be very confusing and might not tell you much,but looking at a video like this how can there still be questions?
Kabylake is ~50% faster in games that's why Ryzen needs the full boost of it's SMT just to,not even,keep up with kaby because there is still a clock difference...
Ryzen needs 50% (that's the gain they have in cine and co from SMT) "more CPU" just to keep up,there is zero future proofing or multitasking on the side ability at all,the 4c/8t get's you less FPS then a 4c kaby even at full load...
Ryzen is still a diamond in the rough, and for gaming the 1600~$200 is worthy of an honorable mention. For other than gaming it is a great 12 thread CPU. It's good enough for gaming at 60HZ for most newer titles with no problems using a 1060/480/580. You might have to adjust quality settings depending on the title, but it's really not that big of a deal IMO. RAM prices are high, and I think these prices help offset the motherboard cost for the i5-8400. Cheap motherboards with cheap 6 phase VRMs are not my style. All in all I think the i5-8400 being the better buy for gaming, and a good all around CPU. When cheaper boards are available it will increase the value of this CPU. Now for some information on Ryzen, and it's potential!
From 2133MHz to 3200MHz, depending on the title, Ryzen gains 7.5% more FPS than Intel from Infinity fabric being tied to the memory clock reducing latency with faster RAM.
AGESA version 1.0.0.4 reduce latency by 6ns.
Overall Ryzen does matched up very well with Intel despite Ryzen having 92% worst memory latency.
Edit: If They fix this problem with the following generation of Ryzen you should see a nice increase in performance in memory bound workloads like gaming.
Memory OC Showdown: Frequency vs. Memory Timings
Posted by rhallock in Gaming on Jul 14, 2017 11:20:08 AM
100% usage in today's games...isn't that what everybody keeps on chanting about regarding the i5s and not being "future proof" because of all the threads that games will use in the future?! (Well, see here,they already do)I don't see how this post is relevant to what Elfear was saying but looking at a screenshot with performance metrics definitely doesn't answer the question whether a CPU would be future-proof or not.
Now that the i3 is 4 real cores anything beyond that is already "enthusiast class",people will want it but nobody really needs it for gaming or general "budget" work.Matter of fact is, neither CPUs are in truly ideal situations for budget builds: the Ryzen chips need fast (and expensive) memory and the i5 8400 needs a more expensive motherboard because there are no other choices right now. The cheaper mobos for the i5 8400 are coming, but will RAM prices come down anytime soon?
I think this fact gets missed a lot when people do value comparisons. Whilst it is often mentioned that there are no cheaper alternatives to Z370 mobos for the i5 8400 (a valid point, may I add, especially for a locked CPU) how many people asking for honourable mentions for the 1600/1600X based on price mentioned that you need (as of today) really expensive DDR4-3200 CL14 RAM just so it can be in the same rough ballpark as a stock 8400?
The quad core 7th gen i5s and 8th gen i3s were considered to be running out of steam because of lower 99th percentile frame times in CPU intensive multithreaded games.100% usage in today's games...isn't that what everybody keeps on chanting about regarding the i5s and not being "future proof" because of all the threads that games will use in the future?! (Well, see here,they already do)
Also the whole previous conversation was about benchmarks at "low" resolutions,well if your CPU is at 100% then that's a good benchmark for comparison,not for real life maybe but for comparison it is.
Now that the i3 is 4 real cores anything beyond that is already "enthusiast class",people will want it but nobody really needs it for gaming or general "budget" work.
The potential also comes as a 'cost', not to mention that most sites already test Ryzen with DDR4-3200 already.
I've said this many times before, and your post illustrates it perfectly - Ryzen *needs* faster memory to be competitive in gaming. And by competitive I mean to be considered a worthy alternative to Intel, because even with DDR4-3200, Ryzen is still slower than an i5 8400, even if the 8400 is running cheaper DDR4-2666 memory
https://www.gamersnexus.net/hwreviews/3086-intel-i5-8400-cpu-review-2666mhz-vs-3200mhz-gaming/page-4
I think this fact gets missed a lot when people do value comparisons. Whilst it is often mentioned that there are no cheaper alternatives to Z370 mobos for the i5 8400 (a valid point, may I add, especially for a locked CPU) how many people asking for honourable mentions for the 1600/1600X based on price mentioned that you need (as of today) really expensive DDR4-3200 CL14 RAM just so it can be in the same rough ballpark as a stock 8400?
Of course, you can also run DDR4-2666 with the Ryzen setup and save $50, but is it worth the peformance penalty? Probably not. Actually, I would like to see this tested - Ryzen 1600/X vs i5 8400 running more budget friendly DDR4-2666. I suspect many AMD fans wouldn't want this scenario tested though, because AMD would lose additional ground in gaming, and they are already behind to begin with.
Matter of fact is, neither CPUs are in truly ideal situations for budget builds: the Ryzen chips need fast (and expensive) memory and the i5 8400 needs a more expensive motherboard because there are no other choices right now. The cheaper mobos for the i5 8400 are coming, but will RAM prices come down anytime soon?
I have compiled the results below in a table. Overall I gained 15% in my fps by taking the memory from 2133MHz to 3200MHz. Interestingly, the gains from 2667MHz to 3200MHz were approximately double those from 2133MHz to 2667MHz, despite an identical 533MHz increase. Does this mean increasing returns as th4e frequency gets higher? The plot thickens.
I'm not familiar with the fabric topology and what the parameters are for this technology, perhaps the stilt would be able to offer some ideas..but would it be possible to either fix the fabric at the speed it would be at ddr4 3400 speeds, or maybe start the fabric frequency scaling much higher than it is at ddr4 2133...perhaps if not fixed then the same speed as memory?This is from a thread I started on the AMD forums, when I noticed the FPS difference from reviews with low speed RAM(2133MHz). It show their isn't a fix rate of increase, but that the faster the RAM the greater the increase if FPS, because the latency is so high. That faster RAM reduces that latency. Same happens in reverse slower RAM make Ryzen perform much slower.
https://community.amd.com/message/2790538#comment-2790538
The biggest thing holding Ryzen back in gaming vs. Intel isn't clock speeds(the i5-8400 3.8-4.0GHz). The deficit comes from Ryzen's nearly double latency times compared to Intel. If they fix the latency issues, even buy cutting them in half, I believe you will see Ryzen starting to beat Intel processors when it comes to FPS. But until then...
I'm not familiar with the fabric topology and what the parameters are for this technology, perhaps the stilt would be able to offer some ideas..but would it be possible to either fix the fabric at the speed it would be at ddr4 3400 speeds, or maybe start the fabric frequency scaling much higher than it is at ddr4 2133...perhaps if not fixed then the same speed as memory?
Perhaps something like this was the original plan, but power consumption or bugs meant they had to have it at half speed of memory controller? I would imagine the nature of the infinity fabric means it's very scalable and ripe for optimisations and adjustments.
Like you said, the main reason for Ryzen underperformance in some games and apps is the latency..to both main memory and inter ccx, if they could increase fabric speed that would massively help.
So would increasing L3 and clock speeds.