Nick Tagataka, 23 Sep 2020Antutu is available on both iOS and Android, but just like Morsel says it's not a cross-p... moreOnce Antutu said that they're not favoured in people compare iOS vs Android. Later on they equal their taste item.. It's more od the hardware benchmark test and obviously a cross platform or or available for both kernals like Metal or Cuda.. Hence you're looking don't know much about it.. Company who caught cheating before would never have caught then.. And for 3d mark any authentic sites didn't anything about cheating with SD device ie. TPU.. So don't hype about anything much..
iOS Never Again, 23 Sep 2020"On lower end Mediatek phones, they found a list of benchmark apps and associated codes t... more"Big deal, you consider that cheating?”
Is there any other way to put it? Those performance boost occurs REGARDLESS of which power mode you're using on your phone RIGHT NOW. The OS detects whether you're using the benchmark apps in the list and overrides the current settings to increase the scores, and most phones that behaved so did it without showing any indications to users that they switched the mode just because they were running benchmarks. You can't be seriously endorsing that kind of attitude from OEMs.
"pushing SOC to its limits in benchmarks is the purpose of the test (to calculate the highest potential score of the SOC"
No, benchmark scores should be the indication of real life performance, which means thermal limitations of devices must be reflected to the results as well. You could claim that amphetamine should be allowed for sports competitions because it would allow athletes to fully utilise the muscles in their bodies without feeling anxious or physically tired, and people want them to perform in their best conditions - in reality, any sort of doping is prohibited in those events. Many of those "performance mode" that triggers during benchmarks not only increases the clock speed of SoCs to the unrealistic levels but also sustain them while disabling the thermal throttling. That's simply unacceptable.
"yet they choose to deliberately ignore such a big news"
Again you have no proof whatsoever that they deliberately "ignored" the news. Emotionally loaded languages are not enough for supporting your arguments.
"Go read Galaxy Note 4 review from all other tech sites vs Galaxy Note 4 review from Anandtech"
I read it and their impression on Note 4 seems to be quite positive overall, with any possible negativities coming from the fact that they analyse the device more deeply than other reviewers. So what's the issue here?
Nick Tagataka, 23 Sep 2020"We are talking about cheating here"
Right, that means you basically ignored what I... more"On lower end Mediatek phones, they found a list of benchmark apps and associated codes that will activate performance mode when running those apps in a file under the vendor folder, and they already confirmed that those phones behave differently when running the same benchmark apps with different package name"
>>> Big deal, you consider that cheating? No wonder they are not being taken seriously.
Performance mode exist on all phones. It can be manually tweaked and unlocked via root.
It's not given to the users our of the box to preserve battery. The latest mi 10 ultra has that mode available for the users out of the box. A few gaming phones have it as well.
At the end of the day, pushing SOC to its limits in benchmarks is the purpose of the test (to calculate the highest potential score of the SOC). Every phone fails in sustained tests to heating and gets reduced scores except gaming phones. Yet Anandtech calls it cheating as their beloved iPhone scores less.
"Apple did that exactly to prevent the devices from not functioning properly, not force their customers to upgrade their phones (You can replace the battery to resolve the issue). What they did was not the real problem but the fact that they didn't tell customers about the deliberate performance throttling was. And you're again mixing up the benchmark "cheating" and a company possibly "cheating" on their customers and playing a word game to make your point"
>>>Nobody is required to pay 500 million unless the scandal is big, yet they choose to deliberately ignore such a big news. You are a good spokesperson for covering up of the pitfalls and shortcomings of Apple. Keep it up, at this rate, you'll be hired by Anandtech soon.
"Moreover, not covering a single topic on their website tells absolutely nothing about the accuracy of their benchmark results (most of which can be confirmed by simply searching for the scores in benchmarks' respective websites). I said it and I'll say it again: "Website A doesn't report news B so A is biased" logic doesn't stand on a solid ground."
>>> Every other site had an article on it but Anandtech conveniently decides to skip it.
No bias there. Good joke!
Go read Galaxy Note 4 review from all other tech sites vs Galaxy Note 4 review from Anandtech. No bias there either right?
Nodal to Mesh, 22 Sep 2020Doesn't matter if they didn't because in gsm test sd phone won in more.. but the thi... moreExactly, all the benchmarks where Apple doesn't do well, is considered unreliable by them or accused of cheating.
Nick Tagataka, 23 Sep 2020You don't read my comment do you? I've been talking about the I N D I V I D U A L pr... moreSpeed test means it not only judges individual app opening times but also takes into account how fast the phone is by opening a bunch of apps in a short time.
This is where iPhones lag behind as they don't have true multitasking.
They also have a RAM management round.
When you open individual apps one by one with a significant gap in between them, iPhone does well. The moment the system is taxed with multiple apps being opened in a short time, iPhone lags behind. This is why phonebuff's tests get different results compared to all other tech channels doing speed tests.
Besides, iPhone isn't faster at video rendering, it depends upon apps.
Certain apps are faster on iPhone, certain apps are faster on Android.
Juan Bagnell has explained how is pixel phone or OnePlus 7t rendered videos faster than iPhone. It really depends on the app.
Heavy Games open at the same time as other Android flagships except for pubg.
iOS Never Again, 20 Sep 2020We are talking about cheating here. There is an article on "supposed" benchmark chea... more"We are talking about cheating here"
Right, that means you basically ignored what I said about Galaxy S4 in my previous comment.
"There is an article on "supposed" benchmark cheating which is unverified at best"
On lower end Mediatek phones, they found a list of benchmark apps and associated codes that will activate performance mode when running those apps in a file under the vendor folder, and they already confirmed that those phones behave differently when running the same benchmark apps with different package name. In the past they also reported benchmark cheating done by Android phones multiple times which were then admitted by UL and GeekBench which resulted removals of those devices from their lists. So not only they found a person who looks to be high on a street, but they also found a drug from his/her house. Those manufacturers have already been found guilty and there's no other way to put it, and bringing that up to question Anandtech's trustworthiness is simply unsightly.
"Yet cheating on batterygate where users will be forced to upgrade"
Apple did that exactly to prevent the devices from not functioning properly, not force their customers to upgrade their phones (You can replace the battery to resolve the issue). What they did was not the real problem but the fact that they didn't tell customers about the deliberate performance throttling was. And you're again mixing up the benchmark "cheating" and a company possibly "cheating" on their customers and playing a word game to make your point.
Moreover, not covering a single topic on their website tells absolutely nothing about the accuracy of their benchmark results (most of which can be confirmed by simply searching for the scores in benchmarks' respective websites). I said it and I'll say it again: "Website A doesn't report news B so A is biased" logic doesn't stand on a solid ground.
iOS Never Again, 20 Sep 20202-3 apps opened with that robotic arm is equivalent to a person opening 5 apps in the same tim... moreYou don't read my comment do you? I've been talking about the I N D I V I D U A L processing/app opening for this entire time, not how many apps you can open in a certain time frame. No matter how fast you can move your finger, Photoshop/Lightroom processes photos in shorter period of time on iPhone. Videos are rendered quicker. Heavier games and office files load faster. And their robot arm is much better at measuring this accurately because there is no delay or at least the delay is consistent when tapping the display with the tip of the stylus. Don't make me repeat the same stuff over and over again.
Nodal to Mesh, 22 Sep 2020Doesn't matter if they didn't because in gsm test sd phone won in more.. but the thi... moreAntutu is available on both iOS and Android, but just like Morsel says it's not a cross-platform benchmark and its code is not written in a same way, hence the results from iDevices and Android devices are not comparable. That's why Antutu shows separate benchmark charts dedicated for Androids only and iDevices only. As for 3DMark, it's the devices with SD SoCs that are suspected of cheating (though this is not confirmed), not the benchmark itself.
Mediatek sux, 22 Sep 2020Looks like you are stuttering even when typing. All i have to say to you is you shouldnt consi... moreAll you logics are fair but you can't compare Apple SoC and Android SoC on AnTutu.
They are written differently. iOS Metal API and Android Open GL ES and Vulkan API.
Even if SD gets away with peak CPU performance with the new SoC it won't be faster in regular usage just because of the 2x monster power core running at same frequency.
Also if Apple goes 1+3+4 configuration we will see Qualcomm falling behind even 2 year old Apple SoC like it used to be 1 year ago.
Nodal to Mesh, 22 Sep 2020As I said you're one Google graduate but nothing else.. I know about tegra x1 before you ... moreLooks like you are stuttering even when typing. All i have to say to you is you shouldnt consider flopcount as the measure for game or benchmark performance. Isa is the machine language to communicate between hw and sw, and by gpu architecture it means how many or which components a gpu has or how they are arranged - these differences are what makes something like amd have high flop numbers, while performing less in games. You can also see these in phone gpu benchmarks where mali performs better than adreno in one test of a certain benchmark, while adreno gets higher fps in some other test of the same becnhmark. Eg gfxbench. While it is true that qualcomm sq1 has 2tflops power, its useless if it doesnt show in real games. If it does, then all is good.
Yes i have checked garys speedtest g where 865 and a13 have similar scores. I dont know if i should trust it because apparently the test runs at native resolution but there was too little difference between 1080p and 1440p phones with the same soc. Maybe he forgot to set it to 1440? Idk.
Just copypasting my earlier comment which you never bothered to think about -
'Qualcomm claims about 950 fp32 gflops for sd855. But when you check gfxbench you will see a12z score 4 times fps on aztek benchmark with only 1300 gflops. On antutu you will see it score double that of sd855 gpu score.
Qualcomm used to score higher fps than apple a few years ago even in gfxbench and anandtech never tried to hide it, anandtech has even did qualcomm sponsored articles. But now they noted qualcomm phones getting unusually hot in 3dmark and thats why they stopped testing with it. In gfxbench and basemark apple scores higher, and they never noted any unusual behavior.
Also im not going to check each and every benchmark on both gsma and anandtech for every phone released just to see if they are apple biased or not. One gaming phone with the latest sd865+ on its highest performance mode with a cooling fan attached still scored lower than 11 pro max, and the only tests they both did gave the same results on both sites. How is that not enough?
You asked for my qualifications right? I guess its having better comprehension skills than you.
Mediatek sux, 21 Sep 2020You are beyond help, anandtech never did gfxbench onscreen test of any kind, only offscreen on... moreDoesn't matter if they didn't because in gsm test sd phone won in more.. but the thing is anand only showed in favour of Apple we know why and then called other test fraud where Apple didn't do well,, lol.. But in other tests site sd won.. Like antutu and 3d mark.. and before you call out 3d mark spy for ch*ating you should have known that it's the of most used benchmark by most of true tech reviewers and reporter ie. GN made a hall of fame in gpu number with 3080 recently.. Even got higher ratings than any other benchmark.. Go get admission in elementary school
Mediatek sux, 21 Sep 2020Sigh, why is it so hard to understand for you? Have you heard of the nvidia tegra x1? An nvidi... moreAs I said you're one Google graduate but nothing else.. I know about tegra x1 before you a smart device.. But the problem is you're merging Nvidia and Amd unknown architecture with arm dedicated isa.. What about Mali then? Yeah I know you're only saying things can be found on internet.. Amd working with Sammy for their latest integration is like one hybrid system isa.. Old Mali, Adreno, Apple or powervr for arm all follow ARM isa because they're not company to develop new isa but designing under isa.. And it's not like Nvidia, amd or radeon gives their own isa to them.. But what benchmark lol? In most of cases Adreno in precision get higher scores.. Both a13 and sd865 plus got very similar precision numbers and go check the stretch speed test between both SOC by Garry where both finished the result very similarly.. Not this far adandpaidtech shown in the benchmark although in 3rd mark sd won.. I know in gfx apple must won. Go visit 3d mark and antutut..
Let me know your qualifications then I might guess.. But as for now you're looking another dude who never admitted in higher level of study..
Nodal to Mesh, 21 Sep 2020Lmao! You're looking another Google graduate who knows nothing but telling whatever read ... moreSigh, why is it so hard to understand for you? Have you heard of the nvidia tegra x1? An nvidia gpu working with an arm cpu and arm isa? Or heard about amd working with samsung for new exynos gpus? Look, the amd and nvidia cards have different architectures, but they both work with the same x86 isa, what is wrong with your mind? Or eyes?
Flops count is not just useless comparing nvidia or amd cards, its useless when comparing every other gpu like adreno, mali, powervr. Seriously why dont you even question why qualcomm and apple have such different benchmark results with no relation whatsoever to flopcounts? Go to the gfxbench site and see for yourself.
Looks like you wasted your time on the "bwm and nvidia joint program", since you still dont know your stuff. And please, dont tell anyone to contact you to learn, let them keep their brains.
Mediatek sux, 21 Sep 2020Its clearly you who knows nothing, read up more on how mali and adreno and powervr are differe... moreLmao! You're looking another Google graduate who knows nothing but telling whatever read randomly on internet.. All arm manufacturers must follow arm instructions in every single chip in the SoC.. Even Intel use x86 isa for their igpu or phi and latest xe graphics processor.. Only amd and nvidia has completely different architectures and isa even top tech site or media group couldn't get those infos easily.. Even I've been telling people that comparing Nvidia and Amd with floating speed is not that helpful for years.. I got the chance to work for bwm and nvidia joint program.. Ever saw a basic electrical book in life? Contact if you want to learn..
Nodal to Mesh, 20 Sep 2020Need to clear your eyes again.. Both site did test Manhattan onscreen also where gsm one doesn... moreYou are beyond help, anandtech never did gfxbench onscreen test of any kind, only offscreen ones. go read up both rog 3 and 11 pro max reviews again.
Nodal to Mesh, 20 Sep 2020They not like amd vs Nvidia gpu with different architectures and isa.. All manufacturers obvio... moreIts clearly you who knows nothing, read up more on how mali and adreno and powervr are different. Even if you dont read up, shouldnt it be obvious when i showed how the flopcount had no relation to performance in gfxbench or antutu?
Nick Tagataka, 20 Sep 2020"And does the phone open the app on autopilot"
For this entire time have you been b... more2-3 apps opened with that robotic arm is equivalent to a person opening 5 apps in the same time frame in real life. This is why the phone buff speed tests are not an accurate representation of real world speed tests which are done with your hands as phones like Apple and Galaxy s20 get more time to catchup whereas in other speed tests they don't get time to catchup eventually lagging behind.
In all other tests the legion phone, Xperia 1 II, Poco F2 pro and others are destroying iPhone in speed tests. I'll trust the rest of the YouTube world over phone buffs robotic gimmick.
Nick Tagataka, 20 Sep 2020Because they don't usually make an article dedicated for those kinds of scandals and rath... moreWe are talking about cheating here. There is an article on "supposed" benchmark cheating which is unverified at best yet no article on massive cheating done by Apple?
Suspect cheating in benchmarks of Android phones comes under "delivering review"?
Yet cheating on batterygate where users will be forced to upgrade doesn't come under
"delivering review", Good logic Nick.
Anandtech simply can't be trusted with exposing the truth about Apple.
I know this for 10 plus years. Plz give better sources.
Nick Tagataka, 20 Sep 2020"SD phones are still ahead in more gfx tests compared to Apple"
https://images.an... moreDamn gsm deleted my reply!
Stop showing pictures from anand when they were affiliated with apple before.. Rather than go check test on gsm where more gfx tests won by sd devices
Not denying Apple better cooler method with built in heat conductor and insulator layer same time upward.. Which is better usual heating pipe for heat speared..
But how silly you're comparing 3gb kiddo's games on iOS or Android platform to 100gb pc games even many available for arm platform windows.. Lmao.. Get me when you finally learn something more informative..