Does anyone feel like posting their benchmarks on the custom ROMs they are running? I know not every GTab will run the same but I would like to get a basic idea of what the hottest scores are. Thanks.
TNT Lite 4.3.x with stock kernel is ~40 fps (Nenamark) and ~2300 in Quadrant. On average.
There are definitely higher scores than that, with some of the other modded ROMs.
2200 with vegan 5.11 Benchmark Atuntu
Thank you guys for posting your benches. I had seen some one some where, posting that they get 60fps. Any one getting 60fps?
there is guide in dev forum where you can get up to 77fps, but the recomended is under 75.
You might mean the frame rate. The default is 50, but TNT Lite and gADAM up the frame rate to 60. At least one mod ups it to 77.
That's not really a benchmark, though - I assumed you meant something like Quadrant.
Im running caulkins rom and clemsyn's kernel #6. I'm averaging around 2400 on quadrant and 3000 on smartbench. The only thing faster on smartbench is overclocked xoom
It looks like most of the new ROMs run pretty fast. If i can get 2200-2500 in Quadrant I will be happy.
I get a little over 3000 qudrant
johnboyjr said:
I get a little over 3000 qudrant
Click to expand...
Click to collapse
what ROM/mods you are running?
i got several scores in the 2400s today and one personal best of 2518.
@johnboyjr
Are you bull****'n? 3000? i havent seen anyone say they are hitting the 3000 mark. I'd like to know what you're running too. Did you overclock? I dont think i want to overclock mine.
nsmartinez79 said:
Im running caulkins rom and clemsyn's kernel #6. I'm averaging around 2400 on quadrant and 3000 on smartbench. The only thing faster on smartbench is overclocked xoom
Click to expand...
Click to collapse
I have the same setup on my gtab... got 39.1 fps in nenamark, and 2399 in quadrant.
I just set my gtab up today so it doesn't have much on it yet.
There is a post in the development section by chengu that has a loopback fix that enhances read/write speeds on the internal SD. This makes the I/O part of quadrant way faster. I have posted on that thread vegan stock 10x average is 2500+. With the loopback fix its a 10x average of 3100 with peaks of 3333. So no its not BS but really how much do quadrant (or any other test) effect daily functionality.? It is great for bragging rights though, especially to xoomers.
Sent from my Chromatic Magic using XDA Premium App
Mantara said:
There is a post in the development section by chengu that has a loopback fix that enhances read/write speeds on the internal SD. This makes the I/O part of quadrant way faster. I have posted on that thread vegan stock 10x average is 2500+. With the loopback fix its a 10x average of 3100 with peaks of 3333. So no its not BS but really how much do quadrant (or any other test) effect daily functionality.? It is great for bragging rights though, especially to xoomers.
Sent from my Chromatic Magic using XDA Premium App
Click to expand...
Click to collapse
Mantara, do you still have the link for loopback fix? I could not find it .
Thanks a lot in advance.
njshkb said:
Mantara, do you still have the link for loopback fix? I could not find it .
Thanks a lot in advance.
Click to expand...
Click to collapse
http://forum.xda-developers.com/showthread.php?t=902426
Fyi: restart your tab 5 or six times, I know I sounds weird but it takes that many to streamline it completely. Let it stabilize after boot and then power down, wait 20 sectonds and and power on. It wil take up to six minutes on first boot and get less each time.
It uses boot time to write changes.
Have fun ;-)
loopback fix like others have said
Is there any down fall to using the loop back fix? Heat issues or failure? Why isnt anyone implementing this into their ROMs?
No adverse side effects that I have noticed. Most people have noticed a slight increese in playing video from internal memory and web browsing cashed sights. Read the thread its full of good info.
Sent from my Chromatic Magic using XDA Premium App
Related
So I've gotten anywhere between 2.5 to 5.1 MFLOPS using various ROMS and have yet to be able to notice something incredibly different.
710...768...806 - What does it matter? What program other than Linpack shows a sizable difference? Sure, maybe things open quicker? What am I missing here?
I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS.
MFLOPS mean jack when there is little way to observe the difference.
Carreno43 said:
So I've gotten anywhere between 2.5 to 5.1 MFLOPS using various ROMS and have yet to be able to notice something incredibly different.
710...768...806 - What does it matter? What program other than Linpack shows a sizable difference? Sure, maybe things open quicker? What am I missing here?
I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS.
MFLOPS mean jack when there is little way to observe the difference.
Click to expand...
Click to collapse
Linpack MFLOPS - measures the floating point performance of your phone.
710...768...806 - refers to CPU frequencies
increasing the CPU frequency should equate to better general-case performance, including things opening quicker as you mention, but also other types of general snappiness like moving between screens and so forth.
"I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS." - This may have less to do with the performance of your phone and more to do with the emulator itself. Emulation is a surprisingly CPU intensive operation, especially if the emulater isn't well written. Rather than looking a ton into overclocking and JIT, etc, maybe you ought to look for a better piece of software.
Yea,
I've tested most emulators. Wish there was an Atari emulator!
Thanks for the response.
Carreno43 said:
Yea,
I've tested most emulators. Wish there was an Atari emulator!
Thanks for the response.
Click to expand...
Click to collapse
I have run roms with 5.1 MFLOPS and now am running a rom that gets 3. I can honestly say I see no difference.
Spencer_Moore said:
I have run roms with 5.1 MFLOPS and now am running a rom that gets 3. I can honestly say I see no difference.
Click to expand...
Click to collapse
I can see a difference... in battery life! Lolz
g00gl3 said:
I can see a difference... in battery life! Lolz
Click to expand...
Click to collapse
Haha Awesome
it looks like to me that everyone is look at the wrong things.
for example:
I am running a Tom that is getting on a average of 4.9 mflops.
I get smoother screen changes....
streaming videos online is so much faster compared to a 3.0 mflop rom. ...
tubetube and other....... websites.
to me everything I do is faster...
I.don't play game on my phone so I don't know how that is.... but everythng else I do is very much faster.
I love high mflop roms...
I have notice about mflops is that it matters about the kernal that u use.
Isn't it true that the MSM7201 in our phones is already overclocked to get to 528mhz as it is? I see a lot of different places saying Qualcomm chips in general are just not worth overclocking... and since our chip is factory overclocked to begin with... just seems like we're pushing the already-pushed here. But the way this board goes crazy for overclocking... it's contradictory. I don't know what to think, cause I've run Linpack myself and gotten ~4.9 with JIT + OC versus ~2.5 without... but I'm with the OP on this one... only difference I'm seeing is my battery draining faster and my phone getting physically hotter.
xatch said:
Isn't it true that the MSM7201 in our phones is already overclocked to get to 528mhz as it is? I see a lot of different places saying Qualcomm chips in general are just not worth overclocking... and since our chip is factory overclocked to begin with... just seems like we're pushing the already-pushed here. But the way this board goes crazy for overclocking... it's contradictory. I don't know what to think, cause I've run Linpack myself and gotten ~4.9 with JIT + OC versus ~2.5 without... but I'm with the OP on this one... only difference I'm seeing is my battery draining faster and my phone getting physically hotter.
Click to expand...
Click to collapse
I have OC and JIT and getting about 5.1 mflops and haven't had worse battery life or a hotter phone. It could be the battery I'm using but meh (got a replacement one that's 2000 mAh) but I got worse battery life on leak 2.1 than with the rom I'm using now that has OC, JIT, LWP, etc. I can go about 8 hours with heavy texting, moderate internet usage and my lwp's running and it only goes to about 65%
so OC and Jit don't make that big of a difference in gameplay?
Sent from my Eris using the XDA mobile application powered by Tapatalk
What the OP and all the respondents are noting is frankly quite typical of what happens when performance tuning focuses on a single benchmark: the results obtained are essentially meaningless for different kinds of activities on the same device.
That's because there's a whole chain of dependencies that are specific to a given task, any number of which could become the rate-limiting factor; and a different task on the device will have a different set of dependencies and therefore different rate-limiting behaviors.
For instance, let's take writing to an SD card as an example: there's really no way that OC'ing will speed that up in a measurable way - because the CPU isn't the rate limiting factor.
That Linpack benchmark measures floating-point performance using a software library (as the Eris has no hardware FP capability). Most of the apps on the phone do very little FP work at all. But, it's not a bad test of CPU speed, because it performs no I/O. It also may not be very memory bandwidth intensive, either (if the problems it works on stays in the uP cache and there are few page faults).
OTOH, a game emulator needs to write to the graphics display (at a minimum) and possibly also do read I/O from flash.
Different task, different results. Sometimes things can be improved by hardware or firmware; sometimes the software itself needs to be improved.
bftb0
im sorry, but could you just answer in plain english
Sent from my Eris using the XDA mobile application powered by Tapatalk
TheSonicEmerald said:
im sorry, but could you just answer in plain english
Sent from my Eris using the XDA mobile application powered by Tapatalk
Click to expand...
Click to collapse
Lima beans bad.
Pork good.
Slow phone bad.
Fast phone good.
bftb0
Thanks for my laugh of the day on that one.
What I'm trying to get at is -
I should be able to play, at the basic level, Sonic or Mario - Without issues.
At the very least
I prefer roms over market games any day (Sonic, Mario, Zelda, DK-Country) and it cripples the phone, at least in my view, that I cannot enjoy the fruits of old games.
Although, I was able to find some old Atari games - which, thankfully, work without stuttering.
Ok, so I just ran a linpack benchmark on my Aria and got the result of 8.903. I decided to save it since that was the highest I got. I went on the greencomputing website to see the best results for the Aria and the highest one was 32.62 mflops. I wonder if that is even possible on the Aria? If so, how do I get that rom/kernel which gets my that amount of speed. the guy had a dell streak rom and was clocked to 998.4 mHz. here is the link --- http://www.greenecomputing.com/apps/linpack/linpack-by-device/
just scroll down to HTC Aria/Liberty.
btw, I am running CM7 on my Aria.
Why do you want that much speed? 806 is fine. Speeds that high would toast the processor. And the phone would get hot so fast.
Sent from my Liberty using XDA App
I know
andrew.cambridge said:
Why do you want that much speed? 806 is fine. Speeds that high would toast the processor. And the phone would get hot so fast.
Sent from my Liberty using XDA App
Click to expand...
Click to collapse
I know but I'm just wondering how that guy even got his phone to fully boot up at that processor speed. Also, I am wondering if his phone exploded after that. Really, How is that speed even possible? Also, if you look at other results in that list, you will see that people got it to work on 864 which is impossible for the CM7 kernel. So that really makes you wonder how...
Moved to General, since it's not a development thread.
JazzyCarr0t said:
I know but I'm just wondering how that guy even got his phone to fully boot up at that processor speed. Also, I am wondering if his phone exploded after that. Really, How is that speed even possible? Also, if you look at other results in that list, you will see that people got it to work on 864 which is impossible for the CM7 kernel. So that really makes you wonder how...
Click to expand...
Click to collapse
Doesn't really make me wonder... Here's the deal, that guy (or girl) has a hobby. His hobby is seeing how high he can get linpack for android to run. To accomplish this he can go about it one of two ways.
Either he took apart his aria, overclocked the hell out of it, added additional cooling and custom built an optimized kernel for 1 purpose - running linpack for android while unstable. (it's not a phone anymore)
A more logical approach would be to run linpack in an emulator, and spoof linpack into thinking it's running on an aria.
Hope this helps clear up some of the confusion.
He must of got into that girls pants he was trying to impress by having a super overclocked aria
Sent from my Liberty using XDA App
Works every time. ^^
LOL gotcha
Sent from my Liberty using XDA App
wow..amazing what a dedicated person
Hi there. Why am i getting only about 1000 quadrant score, while everyone getting somewhat like 1500 on the same roms?
Do i have a different hardware or what?
Different roms will get you different results. Gingerbread roms appear to run around 1000+ while Froyo roms run closer to 1300+. Run it a few times and you should get a better score. Quadrant doesn't really matter too much though. What's important is how the phone feels for you not what a graphic tells you.
Thanks for reply, but the problem is that i'm having different results than people on the same rom.
And does the quadrant score affects on how smooth the games are running?
Do you have a lot of apps installed on your internal memory? The "cleaner" the phone is the more optimized. Junk apps will definitely decrease performance. As for running games, quadrant doesn't affect how smooth your game is running but does show performance measures that could be related to efficiencies of your phone. Theoretically, the higher the quadrant score the better your phone will be at running your games.
What file system are you using...quadrant takes your free space for apps in score consideration..
If your using data2sd and have 2gb for internal(ext2,3,4) then quadrant is going to show a better score than those on rfs.
Sent from my GT-I5800 using Tapatalk
sopdogg said:
Do you have a lot of apps installed on your internal memory?
Click to expand...
Click to collapse
No, i have taken stats from fresh install, also tried different roms, but never got higher than 1000.
rudolf895
No, i am not using data2sd and i guess those people who get higher score than me not using it too.
Build.prop
Need to edit.
refer to reinderro's build thread, which I can't find now.
Sent from my Xperia Arc using XDA Premium App
Litaskull said:
rudolf895
No, i am not using data2sd and i guess those people who get higher score than me not using it too.
Click to expand...
Click to collapse
Oh ok..
But my point was having lesser apps and having more. Internal space will increase the quadrant..
Sent from my GT-I5800 using Tapatalk
Which rom(s) are you using?
FYI - Fresh install of 2.1 will only get you around 500. JIT enabled gets you around 600+, removing additional junk apps will get you 700+. 2.2 Roms will get you over 1000+ easily. Z's Froyo beta 3 got me around 1500+. Others including Gingerbread hover around 900-1300.
I have tried CM7.0.3 J020c and trip's froyo.
In both roms my quadrant is about 1000, but for other people trip's rom giving them about 1400-1500.
I am on wolfbreaks rom..(as per my signature)
The only other thing i do is move all aps to the sd card that can be moved and leave the ones with widgets on my internal memory.
I average about 1450 to 1550 with my best score of 1609.
Run it a few times as well..
On J's rom I run around 900-1200. Never tried running quadrant on Trip's froyo but on his Arc rom I get around 1000-1250. Again, quadrant isn't that big of a deal. Your games will run fine on your phone. Remember, try to have more free internal memory and get rid of junk apps you don't need.
on wolfbreaks rom. i move all apps to the sd card (except widgets)
1300 - 1400
Okay, thank you, guys, quadrant isn't important, but it's kinda wierd that i'm having lower stats than other people.
Litaskull said:
Okay, thank you, guys, quadrant isn't important, but it's kinda wierd that i'm having lower stats than other people.
Click to expand...
Click to collapse
Its not weird at all. Everyone has different apps and settings which affect benchmarks differently. Higher Quadrant scores don't neccessarily translate to better performance either so why does it matter. Also in order to get reliable scores in quadrant you need to run it twice in a row as the OS will kill apps and adjust as the test runs the first time and the second test is always higher and more consistent.
Sent From My *Super Fly X10*
If I (or anyone for that matter!) help you, dont be afraid to press THANKS!
I'm debating on taking back my device due to getting such low stock Quadrant scores. Around 850 stock on average.
What's everyone else getting? Am I the only one with such low scores? I appreciate any feedback!
Benchmarks are worthless in general, Quadrant is incredibly worthless because it doesn't break down the scores and is EASILY gamed.
For example, disabling per-file fsync on an ext4 system gains you 600 points or so instantly.
Enabling Stagefright on Samsungs gets you 500-1000 points but breaks media playback.
I'm not talking about using mods at all though. I'm just saying I'm getting half of what I've seen as the average STOCK 1Ghz score for this device. My 800 @ 1Ghz vs. others 1600-ish @ 1GHz. I just want to know if anyone else is getting around my scores (800ish) without mods.
What other devices?
I get about 1300-1400 stock...
Hi,
I just ran it on my (US/rooted) SGP5, and got 947.
I noticed that on the comparison graph, there's Samsung Galaxy S, that had slightly lower number ...
Jim
P.S. FYI, I used Fast Boot, to kill all processes, just before running the Quadrant test. I'm not sure if that affected anything.
BTW, Quadrant gives HEAVY weighting to file I/O scores - so Voodoo Lagfix-enabled devices and native-ext4 ones will smoke RFS devices.
tcb4 said:
I get about 1300-1400 stock...
Click to expand...
Click to collapse
Hi,
Wow! Do you know how you got such scores? What have you done to your device (and what is it)?
As I posted, I just got 947 on my SGP5. It's stock, other than being rooted.
Jim
Entropy,
Your completely missing my point... Here it is again. "My SGP5 is getting around 850~average scores on Quadrant stock, as in, in the state it was when I took it out of the box. Reviewers and other owners of this device are getting much higher quadrant scores with their STOCK devices as well.
When I overclock to 1.3GHz, I get around 1400 average while other SGP5's get around 2k+ at 1.3GHz. I just want to see how many are also experiencing such low scores as I am.
As I said, many of the reviewers online say that the SGP5 gets around 1400-1600 with Quadrant without any modifications or rooting. I get around 850. Please stop with the "Well Quadrant isn't accurate because, if you mod it like this.. you get this", I'll say it again, My. Device. Is. Stock. It's not modified, I didn't do anything to it to boost the score or lower it.
You're missing my point.
Quadrant is unreliable and inconsistent as hell. It can vary by hundreds of points from run to run, it's easily gamed, it will do weird unpredictable things if you look at it the wrong way. Plain and simple, it sucks and its results are utterly and completely worthless.
You ARE benchmarking with the CPU governor set to performance, right? Benchmarks with the governor in play are even more worthless and variable.
I just saw 600 points difference between two runs with the same kernel - 992 once, 1619 the next.
I've gotten consistent bench marking results every time, consistently low, but consistant.... Do you own a SGP and can say this from experience or just because you have a similar Android device? I actually own a SGP5 and haven't seen the "992 once, 1619 next" at all... I've never gotten above 1619 or ever had a jump of more than 100 on the same frequency.
Unless your testing this for yourself on the device in question, please stop once again...
Yes, I own an SGP 5. I unboxed it last night. I actually ran Quadrant again, 1700 the next time around. Just more evidence that it's crappy and meaningless. Benchmarks in general suck, but Quadrant is especially bad. At least Antutu is fairly consistent AND breaks out scores by category.
It's consistent with every other device I've ever owned - run-to-run variances of a few hundred points.
Note that over time, your scores will go down on this device because RFS sucks and becomes slower as it's used, and as I said, I/O performance is weighted HEAVILY.
RFS sucking is a known problem with old Samsungs, that's why Voodoo Lagfix exists.
Returning a device based on Quadrant scores is stupid on an epic scale given what a horrifically worthless and inconsistent benchmark it is...
Around 1300.
Heh, anyways.
What's Voodoo lagfix? And RTS? If you don't mind me asking, not familiar with those two. And so far it performs pretty well @ 1.5GHz with 2000+ quad scores, downloading Antutu now.
ZaIINN said:
Heh, anyways.
What's Voodoo lagfix? And RTS? If you don't mind me asking, not familiar with those two. And so far it performs pretty well @ 1.5GHz with 2000+ quad scores, downloading Antutu now.
Click to expand...
Click to collapse
RFS is ? File System - it's some special Samsung file system that just kinda sucks.
ext4 is the current standard Android filesystem format, which is MUCH better.
(think of it as being like NTFS vs FAT on a PC)
Voodoo Lagfix is a set of initramfs scripts that automatically convert RFS partitions to ext4, improving read/write performance a lot.
Entropy512 said:
RFS is ? File System - it's some special Samsung file system that just kinda sucks.
ext4 is the current standard Android filesystem format, which is MUCH better.
(think of it as being like NTFS vs FAT on a PC)
Voodoo Lagfix is a set of initramfs scripts that automatically convert RFS partitions to ext4, improving read/write performance a lot.
Click to expand...
Click to collapse
Ah interesting, I looked into the Lagfix; Think it would work the same for the SGP as it does for the galaxy?
ZaIINN said:
Ah interesting, I looked into the Lagfix; Think it would work the same for the SGP as it does for the galaxy?
Click to expand...
Click to collapse
Yes - it's on my todo list once I get CWM to play nice.
So i have cm7 installed on my NT and the performance is just fine. But i heard that overclocking can increase performance and battery life. I dont really care all that much about performance but battery life is really essesntial to me and i wanna max it out. So as an OC'ing newb i would like to hear from you guys what the easiest way to OC is and what a safe OCing speed is once i know how. Id like a speed thats very stable yet still yields up the benefits. Thanks for all the help you guys
These forums truely are amazing
-Silente
Sent from my CM7 Nook Tablet
Well first of all, we would need the proper kernels in the system folders for it work in the first place (which isn't). From what I have read in the Android Dev thread, they are working on it and the would post instructions and details when the time comes.
myl0h said:
Well first of all, we would need the proper kernels in the system folders for it work in the first place (which isn't). From what I have read in the Android Dev thread, they are working on it and the would post instructions and details when the time comes.
Click to expand...
Click to collapse
Thanks, I'll look forward to that
Sent from my CM7 Nook Tablet
SilentStormer said:
So i have cm7 installed on my NT and the performance is just fine. But i heard that overclocking can increase performance and battery life. I dont really care all that much about performance but battery life is really essesntial to me and i wanna max it out. So as an OC'ing newb i would like to hear from you guys what the easiest way to OC is and what a safe OCing speed is once i know how. Id like a speed thats very stable yet still yields up the benefits. Thanks for all the help you guys
These forums truely are amazing
-Silente
Sent from my CM7 Nook Tablet
Click to expand...
Click to collapse
OCing will not increase battery life. In fact, it'll do the opposite, decrease your battery life. I haven't read this anywhere, but that just has to be the case. It's like saying the driving 100 miles uses less gas then driving 30 miles.
theartofbone said:
OCing will not increase battery life. In fact, it'll do the opposite, decrease your battery life. I haven't read this anywhere, but that just has to be the case. It's like saying the driving 100 miles uses less gas then driving 30 miles.
Click to expand...
Click to collapse
I think that the OP is refering more to Underclocking to get more out of his battery life. I cannot comment when the dev team will release this yet.
yea overclocking with an on demand profile should yield better battery life although the kernels probably already has built in settings for screen off and such?
Sent from my sensation 4G
nba1341 said:
yea overclocking with an on demand profile should yield better battery life although the kernels probably already has built in settings for screen off and such?
Sent from my sensation 4G
Click to expand...
Click to collapse
Yah but you can never have long enough battery life!
Sent from my Nook Tablet using xda premium
I too am looking forward to when we can play with OCing this wonderful hardware. I sure the devs will get to it eventually. First we must get a 3.x kernel developed and then ICS/CM9. Once this happens I'm sure there will be plenty of tweaks/scripts to play with not to derail but is it possible(and safe) to play around with build.prop and init.d? I am currently running CM7 Alpha12.
nerdyjim said:
I too am looking forward to when we can play with OCing this wonderful hardware. I sure the devs will get to it eventually. First we must get a 3.x kernel developed and then ICS/CM9. Once this happens I'm sure there will be plenty of tweaks/scripts to play with not to derail but is it possible(and safe) to play around with build.prop and init.d? I am currently running CM7 Alpha12.
Click to expand...
Click to collapse
Yes, It´s time to start OC in NT, for Kindle Fire there are a lot of OC kernels, Not the same for the NT.
Team-B does indeed have an OC in progress. It is in testing now. So far it seems to run very stable at 1188 Mhz (that is the version I am currently testing, altho Goncezilla has ran his to 1250 Mhz stable I believe)
Know-Fear said:
Team-B does indeed have an OC in progress. It is in testing now. So far it seems to run very stable at 1188 Mhz (that is the version I am currently testing, altho Goncezilla has ran his to 1250 Mhz stable I believe)
Click to expand...
Click to collapse
As OP states i am indeed a noob. If i overclock a device will i save battery or burn it faster?
Sent from my Nook Tablet using xda premium
SilentStormer said:
As OP states i am indeed a noob. If i overclock a device will i save battery or burn it faster?
Sent from my Nook Tablet using xda premium
Click to expand...
Click to collapse
Overclocking by itself will typically decrease battery life. Undervolting, on the other hand, has great potential to increase battery life, even when combined with overclocking.
Sent from my SCH-I500 using Tapatalk
ckevinwelch said:
Overclocking by itself will typically decrease battery life. Undervolting, on the other hand, has great potential to increase battery life, even when combined with overclocking.
Sent from my SCH-I500 using Tapatalk
Click to expand...
Click to collapse
Ok. Whats undervolting?
Sent from my Nook Tablet using xda premium
Running the processor on a lower voltage than stock. Many chips may be able to run just fine using less power.
Sent from my Nook Tablet using Tapatalk