ro.sys.hiritsu (Real AnTuTu result)

[fess]

Hello!
I think we all were cheated by AllWinner.

Try to delete from buid.prop this line: ro.sys.hiritsu=30 and you will see next result from antutu:


If you will change to ro.sys.hiritsu=90 rezult will be: 128068.
Translation of hiritsu is:
http://www.kanjijapanese.com/en/dict...nglish/hiritsu

As a result REAL VALUE OF ANTUTU TEST IS ABOUT 36000!

[mitchell4you]

Hmm... nice find
Q:
What firmware are you testing?
What governer are you using?
What device are you using?

Cheers

[fess]

I found it at this forum:

http://4pda.ru/forum/index.php?showtopic=619317&view=findpost&p=37126970

http://4pda.ru/forum/index.php?showtopic=619317&view=findpost&p=37117585

You can find more screenshots from this links.

As you can see:
Draco Telos, stock firmware 19.12.14

[blinkone]

I just tried this on my draco stock rom, ro.sys.hiritsu=95

antutu score of 143430

Divice did get hot though

[no_spam_for_me]

:
As you can see:
Draco Telos, stock firmware 19.12.14

Where can I see it? Not at your picture

Flashed by factory or where do you get it?
If flashed by factory I have to ask Tronsmart why they don't release it????

[no_spam_for_me]

BTW: had you also tried AnTuTu X?

https://play.google.com/store/apps/details?id=com.emobile.flashlight

If you will get the same results, you know this is the "cheating parameter"... if not, maybe it means something different which effects real performance...

[trebor]

The mysteries of hiritsu
Not altered 25 = 53450

#hiritsu = 0 36507
hiritsu = 30, 60730
hiritsu = 35, 68826

Cheers


GFXBench a score of 450 or less Interesting Graphics. Low Score

GFXBench a score of 476 @ # or Zero
GFXBench a score of 476 @30
GFXBench a score of 476 @35

Not much altered there


For what it's worth, AnTuTu 5.6 with PVRMonitor using dovless score 53450
PVRMonitor overlay shows its using all 8 Cores ?
Maybe a handy tool

[krasnoff]

Why you do not believe Tronsmart engineers? These honest people are not interested in cheating!
You see real and absolute fair results!

[fatbasket]

The news appears to be spreading fast if you google it now.

Has everyone really been shafted or is there a good reason for this?

[Finless]

AMLogic S802 did something similar early on as well!!!!

Bob

[no_spam_for_me]

Not only at early on (and if I'm not wrong, minix disabled it sometimes and enabled it sometimes, but I don't know the last state...)

---

Ok, I done some test at
ro.sys.hiritsu=25
and
ro.sys.hiritsu=100
with
AnTuTu X v.5.1
StabilityTest -> Classic Test
StabilityTest -> CPU / GPU Test
directly behind each other off (except a cool-down between 25 and 100)...
I got nearly the same results...
So it seems to only fake AnTuTu... BUT anyway, it's been a long time known that AnTuTu will be cheated also at e.g Amlogic ... (and because of this, everyone is your own fault if he trust AnTuTu... also anyway, some mods build at the same stock rom are faster at handling with less AnTuTu results...)

---

[freakkid]

I am upset because a purchase was made (thru Geekbuying.com) that advertised (and continues to do so to this very day) these inflated test scores based on Antutu testing. I am not implying that Geekbuying was aware of it...but the fact is that these falsified scores continue to be used as a selling point to justify spending the 149.99-199.99 price...for a product that is, in my eyes somewhat equitable to an Orion r28 which has been selling for 99.99. If I had known then, I would had purchased the Orion r28....

[no_spam_for_me]

I think no need to worry...

At my (of cause there are also differences) rk3288:
AnTuTu 5.3 => ~ 37 K
AnTuTu X 5.1 => ~ 35 K (A80 ~ 37,5 K (37,3 + 37,9))


If I have at look at the performance
=> http://www.freaktab.com/showthread.p...l=1#post264156
the A80 is faster than the RK3288 at CPU...
And here the result of CPU / GPU at RK3288

After 5 minutes
cpu passes: 15 (A80 => 15)
gpu passes: 42 (A80 => 45)

All in all the A80 is faster than the RK3288, BUT I also have to say, that the A80 use the same sensitive heat-handling like the S802, and maybe there is some more potential to optimize it (I don't know if the SoC really needs to begin to throttle a this low temperatures and at these steps??? The RK3288 is much more robust at this... Last but not least in my eyes the kernel is to much oriented at Tablet-use instead of optimized for STB (why not a little bit hotter and using more power???))

Btw at AnTuTu Video Tester you will get at the UBOX = 320, at the Draco = 704 (http://www.freaktab.com/showthread.p...u-Video-Tester).

("... Open Hour Chameleon Android media player based on Rockchip RK3288 processor ... got just 263 points..." http://www.cnx-software.com/2014/12/...ty-in-android/ )

The price difference (rk3288/a80) at the 4/32GB isn't so much, but at the 2/16GB a little bit higher...

BTW: At the unfaked S802 systems you will get ~22K (minix), at the faked S802 ~31K ...

[Q100]

---

Ok, I done some test at
ro.sys.hiritsu=25
and
ro.sys.hiritsu=100
with
AnTuTu X v.5.1

Now tryed on Antutu 5.6 at hiritsu = 99



Cool. Today you can achive so many points you want.

[bloodgnome]

Could someone check other A80 Devices ? Like tablets ect. ?

If there is also this kind of cheating then its clear Allwinner --> A**holes and I'm never ever going to buy anything powered by them (Even if a A800 comes out having 1.000.000 Antutu Points).