Gadgetory

welcome back to the hardware unboxing News commonhelp I guess this actually makes two news episodes in a row considering we covered the AMD Athlon 200 GE in a separate video yesterday actually cuts out a bit from today's news corner so any like it's a bit of a shorter one anyway in today's episode we'll be covering some Intel adaptive so you can use benchmark cheating from Huawei and discussing a bit about a recent 20 atti benchmark leak so we're actually going to start with the 20 80 TI benchmark leaks because I want to discuss these in a bit of detail going on the information we know right now about the RT X launch as is the case with pretty much any major hardware launch a couple of weeks before the release date we start to see these sorts of benchmark leaks from websites and YouTube channels no one has ever heard of before fact I reckon a lot of the time these sketchy sources are never heard from after the leaks either anyway what we have here is a Turkish youtuber called PC Hawk Ozzy TV or something like that they briefly published a set of benchmarks comparing the r-tx 20 atti to the gtx 1080 TI in 10 games the video is apparently up for a short period of time and then removed but it was up long enough for websites like 3d Center to collect the results and do a quick comparison now you can see the results here everything looks fairly standard for a comparison we have some numbers in the first column there's a few different numbers in the second column and a few more numbers in the other columns in general this channel is claiming the r-tx 20 80 TI is somewhere between 22 and 55% more powerful than the 1080 TI which i guess it sounds reasonable let's explore the issues with this one firstly Steve watched the video when it was live and immediately noticed that not once do they actually show they had an RT X 20 atti in their hands it would be pretty strange to have access to one of the most anticipated graphics cards this year but then not to actually show it off at all if it was a review accidentally published early you definitely expect to say the physical card in all its glory and if it was a deliberate early leak you'd want to show off the card so everyone can see you actually one which legitimizes the results one of the tests was battlefield 5 and this immediately throw up another red flag this benchmark video was published before the battlefield 5 open beta began on September 4th for pre-order owners and obviously well after the closed alpha in June so to have benchmark that title for this video that have to have exclusive developer access which seems unlikely but the most important thing to note is that as far as we know at this stage most reviewers don't actually have their hands on our TX 2080 or our TX 2080 Ti hardware and this isn't unusual most companies don't ship out their review samples until around a week before the launch that's a pretty typical situation so whenever you see elite performance from more than a week prior to launch you should immediately be suspicious of those results simply because most people don't actually have the hardware of course there doesn't rule out someone actually you know getting their hands on Hardware through other channels but then there's another thing to know companies also don't release their final or near final drivers for their hardware until very close to launch in fact getting a week with the final driver for an NVIDIA GPU would be amazing often it's only a couple of days ahead of time board partners will usually have an engineering driver to test their cards but the final driver complete with optimizations is only available through Nvidia and usually it isn't distributed to partners before reviewers get their hands on it so with any of these leaks this far out from the release it's very unlikely we're seeing results from final hardware with final drivers and most often these sorts of things are just completely fake this leak in particular looks garbage to us and honestly it's not that hard to just make up some numbers there was another leak this week put up by video cards showing a photo of a 3d marked time spire score for the 28th ETI the GPU is listed as an Nvidia graphics device which suggests the use of engineering drivers and in case you're wondering the score of twelve thousand eight hundred and twenty five is about thirty five percent higher than you'd get from a 1080 Ti this is a much more believable leak but as always like to take these things with this guess a grain of salt again there's not a lot of evidence to verify this number and even though it is a photo of a screen with the section redacted it's not hard to fake that sort of thing either the only reason I'm saying this is more believable is it video cards actually has a history of posting accurate information ahead of time but in this case we'll have to wait and see enough on debunking r-tx 2080 TI legs let's look at some real news Michael announced this week but they are the launch partner for invidious r-tx graphics cards providing their GDD r 6 memory for the cards in a press release filled with stock images including this ridiculous wafer vr gaming seriously what is going on in this image Mike once said early efforts with gddr5 and gddr5 X helped to strengthen the relationship and build a highly collaborative effort between Nvidia and Mike on to deliver JD dr6 in lockstep if you might recall micron was the exclusive supplier of gddr5 X for high-end pascale cards micron doesn't state they are an exclusive partner or anything for GD dr6 we already know Samsung also providing the memory for some products in particular Nvidia squad Rho RT X cards but it's likely that both manufacturers will be providing GDD r 6 for a wide range of in video cards considering Nvidia would be one of the largest customer support JD dr6 throwing SK Hynix as well who are reportedly working with Nvidia to intel has reconfirmed that they will be supporting the visa adaptive sync standard in their upcoming GPUs until first announced support for adaptive sync at IDF three years ago but since then there hasn't been any word on when adaptive syncs support was coming or even whether adaptive sync was still in Intel's plans but this week Chris hook confirmed that adaptive sync is still planned for a future generation of Intel GPUs Intel hasn't clarified when adaptive sync will be supported but it won't be for at least another generation the feature is not supported in cable X integrated GPU and that GPU is basically the same in the rest of Intel's 40 nanometer plus plus plus plus plus plus lineup including coffee lake and whiskey Lake I guess at this stage that Intel are planning to support adaptive sync in their next gen 10 nanometer architectures but due to a series of delays with 10 animator that date keeps getting pushed back and also be extremely surprised if Intel's discrete GPU scheduled for launch in 2020 did not include adaptive sync support and yes phaser adaptive sync is basically the same as fruity so when Intel gets around to supporting adaptive sync you'll get variable refresh support on freesync monitors with Intel GPUs this one isn't really PC highway news but it's still pretty funny Huawei have been caught benchmark cheating with their latest phones under both the Huawei and honor brands and we're not just talking about a little bit of benchmark cheating we're talking full-blown major benchmark cheating the sort of stuff they gets you caught embarrassingly red-handed and then tech discovered that phones such as the p20 Pro and honor play were detecting benchmarking applications via the app ID or name and putting the phone into a special secret high performance mode they're significantly boosted the CPU and GPU performance relative to regular everyday usage UL which formally future mark the company that develops 3dmark which is also available for phones found that some huawei phones performed up to 47 percent higher in public benchmarks compared to modified benchmark apps that the phone cannot detect and optimized for and we're talking 47% faster in the exact same benchmark tests this is pretty poor behavior from huawei benchmark cheating may seem like a small issue but it's essentially false advertising as the phone performs significantly differently outside of detectable benchmark apps review sites that use these benchmarks are basically betraying the phones is faster than they actually are in real-world workloads and that's not the fault of the reviewers but obviously a huawei for cheating and it's worse than that as plenty of users download these benchmarks to do a bit of testing themselves to see how their phone stacks up but in the case of these huawei phones the scores they're getting another reflection of real-world game performance on top of that benchmark cheating can paint a smartphone as being much more efficient than it really is for example a phone could be shown to have great battery life in its standard everyday performance mode but then had shown to also have outstanding results in benchmark so you might get the idea it's a great mix of bachelors and performance but in reality the performance you will achieve in apps on battery is nowhere near the level shown in benchmarks hence the better battery life it's not a shock to anyone but cheating in benchmarks is bad and it's also super dodgy anyway in response to this cheating ul banned several Huawei phones from their 3dmark leaderboard for breaking their app detection rules now that's a big deal for huawei as buyers looking at the chance to figure out which phones offer the best performance will no longer see devices like the p20 Pro among genuine top-end contenders like the galaxy note 9 + 1 + 6 however while way says they are planning to provide users with access to the performance mode so they can use the maximum power of their device when they need to final topic for this week ek water blocks have announced a series of new water blocks for NVIDIA Tsar TX 2080 and r-tx 2080 ti at least the reference versions of these cards the GPU water blocks don't come cheap but they will allow those that use open-loop liquid cooling to integrate these cards into their cooling loops and achieve the usual stuff like better temperatures for better over clocks the blocks are full coverage so they cover the diet memory and VRMs there also are several versions available depending on your requirements the cold plate comes in either electrolytic copper or nickel plated electrolytic copper while the cover is available in either transparent Plexiglas or black acetal plastic if you get the nickel plated cold plate version you'll also get integrated RGB strips for extra bling and of course back plates are also available in either black or nickel pricing for the waterboxx range from around 125 us to 140 dollars u.s. depending on the version while the back plates are about 40 bucks that's it for this week's news corner as always if you like this segment and will it in your inbox every Friday click like and subscribe make sure you enable the belt as well for proper notifications consider supporting us on patreon to get access to an exclusive discord chat and I'll catch you next time