Friday, June 16th 2017
Core i9-7900X Skylake-X Review Shows Up
An Intel Core i9-7900X has appeared for a full review at the site Hexus.net. Spoiler alert, it clocks to 4.7 GHz on all ten cores with relative ease (only taking 1.25 V, apparently, though it racked up nearly 100°C in Cinebench at that voltage).
The review praised Intel's overclocking headroom and general muscle in a mostly positive review. Still, not all is rosy in Intel land. They found performance per watt to not have improved much if at all, criticized the high price tag, and Hexus.net had the following to say about the overall experience:
"X299 motherboards don't appear to be quite ready, there are question marks surrounding the Skylake-X processors due later this year, and at the lower end of the Core X spectrum, Kaby Lake-X is nothing short of puzzling."
It would seem AMD is not the only major chip-maker who can have motherboards ill prepared at launch time, even the mighty Intel may have teething issues yet.
You can read the full review (which is mostly positive, by the way) in the source link below.
Oh, and a special shoutout to our own @the54thvoid for discovering this article.
Source:
hexus.net
The review praised Intel's overclocking headroom and general muscle in a mostly positive review. Still, not all is rosy in Intel land. They found performance per watt to not have improved much if at all, criticized the high price tag, and Hexus.net had the following to say about the overall experience:
"X299 motherboards don't appear to be quite ready, there are question marks surrounding the Skylake-X processors due later this year, and at the lower end of the Core X spectrum, Kaby Lake-X is nothing short of puzzling."
It would seem AMD is not the only major chip-maker who can have motherboards ill prepared at launch time, even the mighty Intel may have teething issues yet.
You can read the full review (which is mostly positive, by the way) in the source link below.
Oh, and a special shoutout to our own @the54thvoid for discovering this article.
247 Comments on Core i9-7900X Skylake-X Review Shows Up
They don't bin CPUs based on what temperatures they reach.
They bin then based on what voltage it takes to reach x frequency. They can use this to calculate thermal design power or how much thermal energy in X amount of time with X heatsink installed will be produced.
So changing solder won't allow higher stock clocks, as the chips will still be binned on the exact same way. ;)
In fact the only people a soldered IHS would help are extreme over clockers or bell ends running to many volts through the CPU.
You see this all the time with Ryzen, they're running cool so people keep pumping the volts in not caring about the damage they are doing over the long run.
I don't mean to be rude to you but your level of understand on how the technology works and how companies operate is quite limited, you think you are right because the limited information you "know" would make it right.
But here's an idea, you're ignorant on several important factors and straight up wrong about some of the things you "know".
With intel the bottleneck of TIM is considsrably larger, which is why with solder intel could easily have given the 7900x 3.5ghz base with the same temps or lower, which would help it a lot in benchmarks, and you could get 4.9-5ghz with high-end watercooling instead of 4.7-4.8ghz (and the TIM is still a concern since cheap paste degrades a lot over time unlike nt-h1 for example which will basically last you "forever" and 10-20c drop with mid-range liquid metal means it's definitely no nt-h1 or grizzly kryonaut).
For about the 4th time, a temp drop of 10-15c, which is typical for a delid, will NOT allow you to use less voltage. Now, when you are talking subambient, like when using dry ice or ln2, you are absolutely correct... just not with ambient and that little of a temp change.
If they could have binned these higber, they would have. Paste doesnt matter in this case. 200 mhz on most silicon would have pushed this cpu past their threshold of 140W. Again, lower temps arent dropping power or voltage in the case of a delid.
They paste they use wont fall apart. Stop thinking a story bere or tbere is actually how it is.
God damn this is frustrating having to keep in line the misinformation spewing from your fingers here...
Simply...science and facts you are spitting in the face of here bud.
The second section on that article (under the image) seems to prove my point about the TIM, even they admit that there's a problem and they know way more than me.
I reviewed the cpu myself. I saw 90c at 4.5ghz with all cores running a stress test (aida64 fpu - slightly higher in p95). Games were sitting around 70 iirc. Perfectly acceptable...the processor will lead a long life there, and the tim will be good throghout the life of the cpu.
Again, its decent tim.. does the job just fine at stock, and allows for some overclocking. Sorry its not as much as many would like, but...it isnt shit, and wont crumble in your hands after use as was said by you two in this thread.
Hopefully you two can wrap your head around that and stop being so damn dramatic over it. :)
Again, hotter than "you" like isnt relevant. What the whitepaper says is what you should go by...not personal preference. Well, at least dont hold ypur preferences against it when whitepaper saus otherwise.
You must really hate ryzen 1800x... even with its lower temps, its total overclock is 400 mjz from base to max clock with all cores. 7900x...3.3ghz base to 4.5ghz all cores..seems like it overclocks 3x better...also, 7900x boosts all cores to 4ghz... the 1800x's limit...or clsoe to it. ;)
I still stand by the idea that running that close to the thermal threshold makes me uncomfortable. I tend to monitor my system temperatures closely and 90c would cause me to freak out.
I don't care, nor does intel care about your preference. Your preference is fine... that is your thing, but its OK to run up there as we have all said ad nauseum.
8:55 wraith spire can only handle 1.32-ish volts, not 1.3625 like u12s, which could go a bit further and AIO's can handle 1.4+ volts according to some other reviewers. Heat conductivity and heat dispensing capacity make a huge difference. Considering delidding gives 2 degrees or so lower temps on ryzen 7 and more like 15 on the 7900x, that's essentially still the difference between something like a u12s and 240-280mm AIO, or 0.05v+ max on ryzen, so that's a huge difference in cpu terms. Of course the voltage difference is speculation and you can't compare intel and AMD, but the point is that the limit is moved up 0.05v or so if I'm correct, which is 200-300mhz difference depending on how lucky you get with silicon with 7900x and if you're looking at baseclocks or improved overclocked speeds.
A couple of problems...
1. Software readings are notoriously inaccurate.
2. With #1, a difference of .01v isn't really measureable using software as the dude did in the video.
3. I also addressed that point earlier. Nope... it does not... not in the least. 15C will not allow a noteable drop in voltage. You were good here until the last part. If Im hitting my temp limit Im hitting my temp limit. The TIM used doesn't matter as in the end its still at the temp limit. Now, because it LOWERS temps, you have more headroom. But to state that you can get a higher overclock at the same temp with different tim is a ludacris statement. Did you mean to say that???
Going from something like a wraith stealth to a 280mm radiator, you can drop voltages a fair bit. Of course replacing the TIM isn't quite as big of a difference, but it still easily is big enough of a difference to fall outside of margin of error. Of course I can't proof anything since reviews are done with different chips with better/worse silicon, but I have seen a few trustworthy reviews state the difference between required voltage between air and liquid cooling for example. Just kinda hard to find them when 99% of reviews only uses air or water cooling and it's especially hard to find a review stating voltages required for x frequency. I believe some ryzen reviews did though. I'll see if I can find one of those.