[janesh]

I sold him Nickos, very disappointing performance in daily use. I just saw a pandawill video, where they showed a antutu score above 14000 and quadrant 3750. I think they lie, I just did another quadrant test, result 3050.
Yesterday I looked at a livestream of a skating match from the website of the dutch broadcasting company, the hero ii very very bad, the V812 very smooth.

I think the A31 has potency, I notice some lag, bit more as RK3066, but I think in overall use a bit faster than the RK3066, and video playback is In my view better as Mali 400.

[Nickos-V]

janesh, on 13 January 2013 - 03:11 PM, said:

I sold him Nickos, very disappointing performance in daily use. I just saw a pandawill video, where they showed a antutu score above 14000 and quadrant 3750. I think they lie, I just did another quadrant test, result 3050.

Yeah, that makes no sense because you were getting in the 12,000s for Antutu and only 3050 for Quadrant. Very big difference in #s from PandaWill. Better to get results from non-sellers (ie. Chinese review sites) because you never know if the Chinese sellers are being truthful or not.

A31 stuff I answered in here:
http://www.slatedroi...post__p__493841

[ioncannon]

Quad Cortex-A5 948MHz@1.175V (Typ. MAX), 252MHz@1.0V(Typ. MIN), 1104MHz@1.2V(Benchmark), 256K L2 Cache
32bit DDR3-912
3D Vivante GC1000@480MHz, 2D GPU@642MHz,
video decoder&encoder@214MHz.

[TonyJZX]

i have to be impressed with the level of discussion here

very rarely will you see such a technical dissection of tablet hardware

i am worried about the performance now that you guys mention it

i mean a quad A5 gc1000+ driving a 1,280 x 800 screen is not sufficient to you guys so what am i to expect at 1,920 x 1,200???

will really be interested in the 1st reviews

[astrayred]

Agreed, I have a new found respect for you guys. That's some swell detective work there Tzul!

[F&LSuperstore]

I'm getting consistent low 12000 Antutu from the Hero IIs I have for testing. No more, no less. So that puts it 3000 faster than a Hero I. I'm going to do a side by side benchmark here in a few minutes. And post the stuff on Youtube.

I wrote this http://flsuperstore....gic-aml8726-mx/ on December 6th.

I was pretty much going on number calculations from a developer contact I have at Amlogic. So I guess in the end the guy was right as suspected.

I gave these ATM7029 tablets a fair chance. So far I'm not impressed.

[Nickos-V]

ioncannon, on 16 January 2013 - 01:05 PM, said:

Quad Cortex-A5 948MHz@1.175V (Typ. MAX), 252MHz@1.0V(Typ. MIN), 1104MHz@1.2V(Benchmark), 256K L2 Cache
32bit DDR3-912
3D Vivante GC1000@480MHz, 2D GPU@642MHz,
video decoder&encoder@214MHz.

How did you get that info? Was it from an app? Or from terminal command? Let me know.

F&LSuperstore, on 24 January 2013 - 05:19 PM, said:

I'm getting consistent low 12000 Antutu from the Hero IIs I have for testing. No more, no less. So that puts it 3000 faster than a Hero I.

I gave these ATM7029 tablets a fair chance. So far I'm not impressed.

Antutu is bad benchmark to use. Do not trust it. ATM7029 ranks very well in Antutu but does badly in bunch of other benchmarks like Vellamo's HTML5 & Metal, Nenamark2, Quadrant, etc.

No idea how Antutu gives it such a high score when other benchmarks do not. I am disappointed that Antutu ranks it very high. Gives false illusion that it is really great when it is not.

The better ARM chips today, ranked in terms of GPU performance are: A31, Exynos Quad, RK3066
Add to that RK3188 once it comes out in about 2 (or 3) months time. Those are the only really great chips coming out of China to choose from.

In terms of battery life: RK3066 then Exynos Quad/A31. Difference of about 30-60 mins on 8" 1024x768.

PS, ATM7029 compares to Tegra2/OMAP4 GPU performance. The other ARM chips I list above perform equal or better than Tegra3 GPU. 8726-MX performs in middle of Tegra 2&3.

[Mjhieu]

After read from page 1 to here. I decide to witch to Nexus 7 and Nexus 4, enough to know more Chinese tablets. I pay for what the best I should deserve. Chinese stuff is not good anymore.

[nyxuu]

I read all you wrote, but not sure I understand.
Does that mean that Rockchip RK3066, 1.6GHz, Cortex A9 dual core is better choice than ATM7029, Cortex A9 Quad core 1.5GHz, Vivante GC1000 GPU.
I'm in doubt between these :
Teclast P88 Dual Core 16GB Tablet and 7" Ainol NOVO7 Venus Myth Quad Core
What do you recommend?

[Tzul]

nyxuu, on 30 January 2013 - 11:19 AM, said:

Does that mean that Rockchip RK3066, 1.6GHz, Cortex A9 dual core is better choice than ATM7029, Cortex A9 Quad core 1.5GHz, Vivante GC1000 GPU.

Yes, that's what it means.
A Rockchip RK3066 will outperform an ATM7029 in most use cases, because it has a faster CPU (even though it's just dual core) and a better GPU.

However, it's your decision.
If you like the Venus better for some reason, then go for it.
Just be aware that even though it's sold as a Cortex-A9, it's actually Cortex-A5. And that the Vivante GC1000 is rather weak and has compatibility problems in some games.

[nyxuu]

Thank you for the fast answer. In fact I like better the Teclast p88. I'm in love with it, but i am scared that is not so famous brand. And I didn't found many reviews

[F&LSuperstore]

After much trial and error and thousands of dollars wasted, I would only recommend Ainol, Zenithink, Onda, Cube, Ramos & Eken. Eken is kind of on my almost bad side, but their return rates have been on the low side for "most" of their products. Which is why I will still recommend them, but if they put out another bad product that goes through my company I will be done with them forever and let people know they should do the same.

Ramos is also one that I'm kind of dubious about. Their W22s had a 50%+ touch screen fail rate. But I never had one return of their W17s. So you have really good and really bad....

Avoid at all costs Sanei & Ampe. And also avoid any tablet out that you can't easily find the firmware for on the net.

[nyxuu]

F&LSuperstore , Thank you very much. I will focus on the Ainol then - more sure thing )

[Tzul]

Nickos-V, on 24 January 2013 - 11:16 PM, said:

No idea how Antutu gives it such a high score when other benchmarks do not. I am disappointed that Antutu ranks it very high. Gives false illusion that it is really great when it is not.

So, Antutu Benchmark was just updated to version 3.1.1.
And I've been told that the Hero 2's score dropped from 12000 to 6200 or so. Can somebody confirm?
My Aurora 2 (dual-core A9) still gets 9200.

Maybe Antutu fixed whatever caused the ATM7029 to score so high in RAM/CPU. Unless the new Antutu version actually has a bug there. But I doubt that, as it doesn't contradict other benchmarks anymore.

This post has been edited by Tzul: 01 February 2013 - 01:54 PM

[Jote]

Couple days ago we ran few simple tests on a other forum comparing Amlogic 8726-mx (Hero/Crystal) and Actions ATM7029 (Hero2). Linpack was used to compare pure calculation power of the CPU by doing only floating point calculations.

https://play.google....g.linpack&hl=en

(Test by me)
Samsung Galaxy S3 overclocked to 1.5GHz pulls 146.158 MFLOPS in 1.15 seconds. This is a quad core Cortex A9 reference with a overclocking to ease comparison to ATM7029.
Crystal@1.2GHz 67.874 Mflops 2.48s


(Tests by second guy)
Hero@1.2GHz
Single threaded:
Hero 1 51MFLOPS 1.63s
Hero 2 30MFLOPS 2.77s
Multithreaded:
Hero 1 79MFLOPS 2.13s
Hero 2 69MLOPS 2.42s

(Tests by third guy)
Hero 1 with CM10.1 overclocked to 1.5GHz
Single 51,941MFLOPS 1,61s
Multi 85,185MFLOPS 1,98s

Test is very short and small disruption can cause lots of fluctuation but comparing A9 dual core and quad core, the tests seems to be in line. Amlogic 8726-mx tests seem to be in line with all participants.

Cortex A5/ATM7029 seems to be absolute disappointment. Stock rom from Ainol seems to be ok though little bit can be forgiven, but by performance the chip is a step backwards. They sell sh*t and call it ice cream.

This post has been edited by Jote: 01 February 2013 - 09:05 PM

[Nickos-V]

Tzul, on 01 February 2013 - 01:44 PM, said:

So, Antutu Benchmark was just updated to version 3.1.1.
And I've been told that the Hero 2's score dropped from 12000 to 6200 or so. Can somebody confirm?
My Aurora 2 (dual-core A9) still gets 9200.

According to this site:
http://mp3.zol.com.cn/352/3529308.html

Antutu 3.1
ATM7029: 6600
A31: 12000
RK3188: 18000

My N80 (RK3066): 11900

Exynos 4412 gets 15xxx with Antutu 3.0(?). Probably close or same with Antutu 3.1.

Seems like Antutu finally fixed their benchmark software. 6600 is way more realistic for ATM7029. ie. gets low total score because of GPU & CPU. Hopefully now the scores are all right.

[sikke]

Ainol new firmwares for Hero ii and Venus versions 0131 have Antutu blocked. Antutu will crash at first seconds of the test. Cheaters!

Old roms give antutu around 6k eaven under. 6.6k maybe with charger connected. I have both hero 2 and venus.

Sent from my JY-G2 using Tapatalk 2

This post has been edited by sikke: 04 February 2013 - 08:03 PM

[Tzul]

Update on the current situation:

It looks like Actions / Ainol removed the "CPU part" line from /proc/cpuinfo in their latest firmwares!
A bit late for cover up tactics, makes them only look worse.

Some background on this:
ARM CPUs have a couple of ID registers, the CPU part number (c05 for Cortex-A5, c07 for Cortex-A7, etc.) is given by the Main ID Register (MIDR).
According to ARM's documentation, the MIDR is "only accessible in privileged modes", which probably means that only the kernel can read it, Android apps can't (maybe with root?).
So, that's why there's /proc/cpuinfo, a virtual file used as an interface to query the CPU info from the kernel. The source code responsible for generating the file's content is located in /arch/arm/kernel/setup.c which uses the "read_cpuid_id()" function from /arch/arm/include/asm/cputype.h to get the MIDR value.

Here's the relevant code from Amlogic's kernel source:

seq_printf(m, "CPU part\t: 0x%03x\n", (read_cpuid_id() >> 4) & 0xfff);

Actions most likely removed / commented out this part. Once they release the source code for the Hero II, Venus, etc., custom firmware developers should definitely take a look at this and undo the tampering.


PS:
An admin on Ainol's official forum has written this related post. Is anyone here fluent enough in Chinese and English to give a summary? Machine translation isn't quite good enough yet.

[F&LSuperstore]

+1s for everyone. Great discussion. I'm glad that Ainol is going toward the A31 on some of their higher end stuff coming out this quarter, but I'm still going to carry their Amlogic tablets till the quad-cores come out because as I've said all along they would be competitive with the newer chips out (Not the quad-core rockchip) but sure enough on the A31 and the ATM7029.

I'm just waiting patiently for Amlogic's quad-core. Hopefully they get me a test run out so I can do some benchmarks and mess around with it.

[Jiefurui]

Just bought the Venus/Myth here in China for about $109. It comes with 1GB RAM, 16GB internal storage and no GPS and bluetooth. I ran the Antutu benchmark test (v3.1) and initially it scored 12000ish using ROM version 131. I just downloaded the 203 ROM and I'll also get the Antutu 3.1.1 and then run the tests again. Antut reports it as having a quad core ARMv7 processor (VFPv4, Neon) and the GPU as GC1000+MP Graphics Engine.

At the time I bought it I had a choice between Onda 811 and Ramos. I wanted a 7 inch model so it's easy to carry around. The Myth had a nicer screen, Onda was an inch model and Ramos didn't have a quad core. Besides, I won't be using it much for 3D gaming but for presentation and also I already have bought a case with keyboard when I previously bought a Momo7. Disappointed with the latter as the manufacturer refused to update the ROM and it died out on me after a year.

Right now I'm having a problem with the computer not recognizing the tablet once it has been connected so I could update the firmware. Anyone knows how it should be done?