Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Latency in cycles hasn't changed. E.g. a 12 cycle L2 cache is already in the Zen cores as demonstrated in #152. So a 12 cycle L2 is not something new to Zen+ cores as someone else pretended. Latency in ns is what changes because the new chip runs at higher clocks!
Even the ancient Sins of a solar Empire got a multithreaded patch last year.
forza 7 i definitely know is set to use 7 threads on both pc and console. then you have all the battlefield games which scale past 4 threads but not 8. i've seen a few other games that i can't think of off the top of my head because they weren't games i cared about that showed gains with 8 thread cpu's vs 4 core/thread cpu's. i'll have to look for them later this afternoon after i get some sleep since i just got finished with a 12h shift.
Let me guess, still sandy-bridge IPC?
14.634% excluding all >= 256-bit workloads, 23.632% excluding single extremities (low / high), 28.002% including all workloads in the suite (30 workloads).
I'd assumed that attachment was the latency others had been referring to, and I'd assumed that it was improvements due to the new process or AMD addressing hardware issues early adopters identified first go around. With the percentages seemingly not aligning with the clock increases (or each other), I'm not sure how it would relate to simply higher clocks... but I'm not going to pretend I understand what exactly goes on at that hardware level and I'd be happy to be educated there.
There was a clock for clock comparison somewhere (still preliminary/preview/leak, so take it with a *huge* grain of salt) that suggested even at equal clocks, Zen+ has lower latency in ns. Improvements in the IMC, perhaps? I understand that the number of cache cycles hasn't changed, but that doesn't mean latency can't be improved elsewhere.
Yes, the memory controller is improved in latency (at expense of bandwidth).
Also "equal clocks" is relative. There are clocks domains aren't accessible to final user. You can clock equal only the accessible clocks, not the not accessible (which are seted at fabric).
IPC? With Meltdown/Spectre patches included Skylake IPC is about 15--28% higher than Zen.
Let us ignore outliers and 256/512bit workloads. The IPC gap is 15%. So if Zen+ brings ~3% higher IPC according to AMD, The gap is reduced to about 12%.
IPC? With Meltdown/Spectre patches included Skylake IPC is about 15--28% higher than Zen.
Let us ignore outliers and 256/512bit workloads. The IPC gap is 15%. So if Zen+ brings ~3% higher IPC according to AMD, The gap is reduced to about 12%.
I have 2700X on-hand testing against my 8700K.. you're gonna be surprised when the reviews come out.
why keep it a secret? spill the beans lol
I think his "on-hand testing" is playing games at AMD's testing booth yesterday.
AMD have systems with Ryzen 7 2700X and systems with Core i7-8700K so you can compare games on the different systems.
Spoiler alert from [H] testing... gamers can't tell the difference between systems.
In one game.
To be Honest once Kyle reviews the 2700k. Two things will happen. He will be called Biased against AMD, Or will be called Biased against Intel. Either way its fucking retarded what people will say over a very honest review.
To be Honest once Kyle reviews the 2700k. Two things will happen. He will be called Biased against AMD, Or will be called Biased against Intel. Either way its fucking retarded what people will say over a very honest review.
I think your stretching it a bit. There may be a few drama queens on here but most will be happy to have a review from someone they trust.
I would assume 12AM EST as the NDA takes effect in the timezone it is contracted and signed in. Most Lawyers for Wall Street are NYC.What time does the NDA lift? When the clock strikes midnight on the 19th?
I would assume 12AM EST as the NDA takes effect in the timezone it is contracted and signed in. Most Lawyers for Wall Street are NYC.
What time does the NDA lift? When the clock strikes midnight on the 19th?
So, if I find a forum topic discussing that the moon is made out of cheese, and someone states the latest tests show that it is made of cheese, it makes it indisputable fact? Thanks for removing all doubt why many people have blocked you, as you linked nothing that supports your claim other than a similar discussion in another forum, that is going on here. Makes a person wonder if you just linked your own comment under a different name.
If you want people to listen and entertain your statements, give us some proof other than he said she said type of stuff. Give us some real proof instead of this crap.
I have 2700X on-hand testing against my 8700K.. you're gonna be surprised when the reviews come out.
I'm on record as being against either test because of flawed methodology and even Kyle acknowledged that it is being extrapolated to prove more than it really does. Not to mention that you're now lumping all gamers into a category where the sample size was and is statistically insignificant.
At this point, you might be left feeling disillusioned when considering AMD’s tech demos. Keep in mind that most of the charts leaked and created by AMD revolved around Cinebench, which is not a gaming workload. When there were gaming workloads, AMD inflated their numbers by doing a few things:
In the Sniper Elite demo, AMD frequently looked at the skybox when reloading, and often kept more of the skybox in the frustum than on the side-by-side Intel processor. A skybox has no geometry, which is what loads a CPU with draw calls, and so it’ll inflate the framerate by nature of testing with chaotically conducted methodology. As for the Battlefield 1 benchmarks, AMD also conducted using chaotic methods wherein the AMD CPU would zoom / look at different intervals than the Intel CPU, making it effectively impossible to compare the two head-to-head.
And, most importantly, all of these demos were run at 4K resolution. That creates a GPU bottleneck, meaning we are no longer observing true CPU performance. The analog would be to benchmark all GPUs at 720p, then declare they are equal (by way of tester-created CPU bottlenecks). There’s an argument to be made that low-end performance doesn’t matter if you’re stuck on the GPU, but that’s a bad argument: You don’t buy a worse-performing product for more money, especially when GPU upgrades will eventually out those limitations as bottlenecks external to the CPU vanish.
April 19, 2018 at 9:00 AM Eastern Time
That's when we are posting our reviews.
To be Honest once Kyle reviews the 2700k. Two things will happen. He will be called Biased against AMD, Or will be called Biased against Intel. Either way its fucking retarded what people will say over a very honest review.
There will be so many reviews on different sites that one can draw their own conclusions. I cant wait to upgrade to the 2700x but then they will release a 2800x in like 3 months lol
I don't think they will release a 2800 this time. I have a feeling that they decided that anything above a 2700 will just go to the Threadripper platform.
I was thinking the same thing and to be honest I don't need more than 8 cores. If my current rig wasn't acting strange I would be on Haswell for another 3-5 years.
I went ahead and preordered the 2700x last night. AMD has an absolutely terrible track record with launches but I like the thrill.
same problem when i got my 1700x, they had the over priced MSI board and the carbon was no where to be found when i tried looking. Settled for a crappy asrock killer x370 and could never get memory above 2933, had no problem on the MSI board with 3200 with the same trident z rgbGetting the processor was easy last launch, getting a x370 motherboard was a pain in the butt tho. I would imagine it would be smoother this time.
I was thinking the same thing and to be honest I don't need more than 8 cores. If my current rig wasn't acting strange I would be on Haswell for another 3-5 years.
I went ahead and preordered the 2700x last night. AMD has an absolutely terrible track record with launches but I like the thrill.