• After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.

***Official Galaxy Nexus Pre-Release speculation thread**

Status
Not open for further replies.
Zero shutter lag is going to be a combination of hardware and software. And from the videos I've seen, there's no lag between shots either, so I'm not sure what you're referring to there.

Overall though, I agree. They should have bumped it to 8MP just to match in a comparison matrix. I've gotta assume there's a reason they went this way, as the HTC Rezound is advertising zero shutter lag too, and it has an 8MP camera.

IIRC, the Source had an early version of the nexus that had an 8mp camera. Given all the speculation, I wouldn't be surprised if I misremember this though, so no promises. Assuming my memory is correct there would be a reason why they went with the 5mp vs the 8mp. I would like to think that they have enough smart people to understand the marketing between the 5 and the 8mp cameras.
 
While I can't ever say that I don't want him to just come out and say things most of the time, he does have 3 wonderful kids and a wife to take care of so I don't push. He knows very well that stuff he does say to me will more than likely end up on here.

I've had to sign a few NDAs in the past, a few with Google (search engine stuff and Google Voice stuff, not Android related), and I can't say that I would have broken them either.


Well, if you ever sign one again, you know where to find me. ;)
 
from all the reports i have seen, they went with a nice sensor at 5mp camera because there wasnt the hardware available to read/write 8mp photos fast enough to achieve the "zero shutter leg" effect they were trying to achieve
 
I believe the 1.5GHz Snapdragon is spec'd that way, while the 1.2GHz OMAP is underachieving.

But in pure tests, a 1.2GHz OMAP will be faster than a 1.5GHz Snapdragon, all else equal

I'll be interested in seeing the S3 vs the 4460 with CF-Bench and Nenamark 2 under these speed conditions.

The S3 already slightly exceeds the Exynos at in CF-Bench under those conditions.

While that establishes that the S3 is 20 to 25% clock for clock less effective than that particular A9, the bottom line is that the new phones have these various clock values.

In practical terms, that means that an S3 under these selection conditions is better characterized as viable for a great many users, as opposed to an outright dog.

As for graphics and overall performance, I will watching with great interest how the actual end products comport themselves.

Should be very interesting! :)

PS - I wasn't clear what a "pure test" was in the context, so I replied based on my best guess.
 
I'll be interested in seeing the S3 vs the 4460 with CF-Bench and Nenamark 2 under these speed conditions.

The S3 already slightly exceeds the Exynos at in CF-Bench under those conditions.

While that establishes that the S3 is 20 to 25% clock for clock less effective than that particular A9, the bottom line is that the new phones have these various clock values.

In practical terms, that means that an S3 under these selection conditions is better characterized as viable for a great many users, as opposed to an outright dog.

As for graphics and overall performance, I will watching with great interest how the actual end products comport themselves.

Should be very interesting! :)
Agreed. They've come a long way with S3. Probably will be very close, but either way the results should be so close that differences should basically be noise.

End result is what should matter.
 
Agreed. They've come a long way with S3. Probably will be very close, but either way the results should be so close that differences should basically be noise.

End result is what should matter.

You know me - I only like benchmarks that relate to or predict user experiences.

And I will bet that will be hard to get to with ICS.

Even if the benchmarks put the various phones within striking distance of each other, I think the new task management in ICS will make a noticeable difference.

While I contend the Rezound is a viable choice for a good phone from early reviews and feedback, the Gnex owners are going to see that task management improvement immediately, while Rezound and RAZR users will have to wait a few months to enjoy that new advantage.

Having built SMP OSes, I see a lot of promise in the ICS approach.
 
You know me - I only like benchmarks that relate to or predict user experiences.

And I will bet that will be hard to get to with ICS.

Even if the benchmarks put the various phones within striking distance of each other, I think the new task management in ICS will make a noticeable difference.

While I contend the Rezound is a viable choice for a good phone from early reviews and feedback, the Gnex owners are going to see that task management improvement immediately, while Rezound and RAZR users will have to wait a few months to enjoy that new advantage.

Having built SMP OSes, I see a lot of promise in the ICS approach.
Eh, I'm sticking with Quadrant. Best benchmark out there. :D

Ok, obviously not. And I agree, that ICS really changes the game. It will be interesting. The only benchmark I've seen so far has been this one, indicating performance will be very close to the SGSII.

antutu3.jpg


Some benchmark results from the Rezound (from web):
AnTuTu, average of 3 runs: 5644,5724,5245; 5537.7.
 
You can lead a horse to water, but you can't make him drink. Most consumers don't want to know too much, just enough. I told the story several thousands of posts back about a guy who was convinced his iPhone4 was 4G, because otherwise why would apple call it the iPhone FOUR.

i was cracking up when you told that story
 
Good night Nexi. One post closer to 30,000.

Off to Tallahassee...hoping a new phone will be waiting on my doorstep when I return.
 
Eh, I'm sticking with Quadrant. Best benchmark out there. :D

Ok, obviously not. And I agree, that ICS really changes the game. It will be interesting. The only benchmark I've seen so far has been this one, indicating performance will be very close to the SGSII.

antutu3.jpg


Some benchmark results from the Rezound (from web):
AnTuTu, average of 3 runs: 5644,5724,5245; 5537.7.

Now that I am back off the floor from the Quadrant assessment, I can reply. :)

I think I have thanked that graphic every time you posted it, but I want to say why.

I see six general groups -

  • Pre-superphone
  • 2 groups for early and later superphones
  • 2 groups for earlier and current dual cores
  • First quad core

It substantiates my expectations for the equivalency of the later 2 A9s, but do we know which of the 3 processors the SGS2 had there? I presume Exynos, but can't know that without support.

And the Rezound coming in just below those tracks with my 3D graphics claims for the S3. Still puts it in the right neighborhood.

Main thing I would like to see personally - Vellamo Web Browser benchmark to compare SGS / 2.3 vs the Gnex.

PS -lol, I was thinking of another thread. The S3 is behind the A9s in 3D performance, that was my claim.
 
Just want to say because I haven't seen it mentioned in quite a few posts, but come Tuesday if VZW stores put up all that marketing material they were supposed to get then we should be certain of a 17th release. [based on that leaked screenshot of 15/17th respectively)

I guess my point is I know the rumored date may go back and forth but I would just be happy to get confirmation of a date at this point even if it is later than what we expect, just so I could have something solid to hold on to, and that confirmation may come Tuesday. Yay.
 
Just want to say because I haven't seen it mentioned in quite a few posts, but come Tuesday if VZW stores put up all that marketing material they were supposed to get then we should be certain of a 17th release. [based on that leaked screenshot of 15/17th respectively)

I guess my point is I know the rumored date may go back and forth but I would just be happy to get confirmation of a date at this point even if it is later than what we expect, just so I could have something solid to hold on to, and that confirmation may come Tuesday. Yay.
I'm hoping you're right. That would be awesome if the marketing material is up in the stores Tuesday. I'm out of the country so I'm counting on you guys to let me know what's happening back there.

Speaking of which, time for me to jump on a plane and head to Hong Kong. Y'all have a good night!
 
I don't have a link available, but Andy Rubin has said he's not a proponent of SD cards. It's not a coincidence that the last two nexus phones haven't had one.

wonder why? last couple were back in the days of unlimited data. he should realize the carriers are now limiting data and change his views accordingly

Yes, performance could be a reason. Built-in cards are typically eMMC. Typically, these support hundreds to over a thousand Random Write IOPS (I/O operations per second). Typical SD cards, including Premium cards, on the other hand, have far lower Random Write IOPS numbers--in most cases, under 10.

On top of this, eMMC implementations often support higher data clock rates than external cards, which translates directly into better throughput. Typical values are 10 to 20 percent higher, in fact.

On top of all of the above, often the built-in eMMC is Class 6 or higher (measurement of speed), whereas the external SD card the carrier throws in for free is only class 2. Even if you put a class 10 card in yourself, however, in most cases it will still underperform the built-in speed.

I think Andy Rubin is looking to improve the overall experience of the phone.

Based on what I have seen and experienced in terms of access speed, internal memory is quite a bit faster. I can say without a doubt that my gf's Nexus S (EMMC) pulls up the gallery overview at least twice as fast as my DInc with 8gb class 2 SD card. Not only that, but the time it takes to access files, delete pictures, etc is quite a bit faster on her S.

Certainly a higher class SD card would perform better than a class 2 in write speeds, but AFAIK, in all aspects, EMMC is significantly quicker in my real world, subjective observations.
 
I think Andy Rubin is looking to improve the overall experience of the phone.

Based on what I have seen and experienced in terms of access speed, internal memory is quite a bit faster. I can say without a doubt that my gf's Nexus S (EMMC) pulls up the gallery overview at least twice as fast as my DInc with 8gb class 2 SD card. Not only that, but the time it takes to access files, delete pictures, etc is quite a bit faster on her S.

Certainly a higher class SD card would perform better than a class 2 in write speeds, but AFAIK, in all aspects, EMMC is significantly quicker in my real world, subjective observations.

shouldn't we be comparing external vs cloud? not internal vs external?
 
your point is well taken, though I guess rezound could find one in that bin. granted they're from different manufacturers.



to dispel someone's argument, state their argument, then take it to the extreme. it sounds absurd in the extreme, therefore their argument was flawed from the outset. Right? well, not quite. this strategy is called the "slippery slope argument." As it turns out every argument becomes absurd in the extreme. Doesn't mean that every argument doesn't hold water because it can't survive the slippery slope test.

I'm talking about 1.5 vs 1.2, not i7 vs 1.2. I understand what you're trying to say about design strategy, and glad that you recognize that compromizes have to be made. That's fine. But let's acknowledge them as compromises instead of saying the gnex can do no wrong. They probably could have found a compromise with a 1.5 or an 8mp camera without losing to much on the other end.




you're probably right. what do you think the reason was? batt life?



the look at some, just not indepth. i've heard plenty of average consumers touting the megapixels of their point and shoots. Or that they "bought an LED tv not an LCD." and other fallacies like that. they know enough to get themselves in trouble.



points well taken. good discussion


I've never done much design, and when I did do design, it was never hardware, but......

Isn't it just common sense that if one of the goals of the Nexus is to be a dev platform for ICS that you would not go all "bleeding edge" on the hardware? Wouldn't you WANT a HW platform that is between "top of the line" and "middle of the road" as far as premuium smartphones go?

Back in my software development days, we never designed for bleeding edge hardware....... well, we designed for what was bleeding edge 12-18 months ago (but we weren't writing games ;) ).

Of course, smartphone hardware seems to be running way ahead of Moore's law at this point as well - forget 6 months, it seems like six WEEKS now!

Just my 2 cents worth, YMMV.
 
Status
Not open for further replies.
Back
Top Bottom