Samsung Galaxy Gear Smartwatch - Preview on AnandTech

[Crono]

http://www.anandtech.com/show/7273/the-galaxy-gear-preview-samsungs-first-wearable

The display looks better than I thought it would, but all of the smartwatches seem to have first generation limitations that would prevent me from getting one anytime soon. Battery life is a big one.

If I'm right about the SoC/core and the battery voltage, under extended load it'd be possible to kill the Gear in a little over an hour. You probably won't be browsing the web on your Galaxy Gear so that's probably an unrealistic scenario, but even if we're talking about 5 - 10 hours of use the Galaxy Gear is still more computing device strapped to your wrist than a watch.

That doesn't sound too good and why e-ink might have been a better choice, even if it wouldn't be color or pretty.

 

[TheStu]

I can't figure out why it has a camera, especially if it is a complementary device to your smartphone.

 

[Crono]

I can't figure out why it has a camera, especially if it is a complementary device to your smartphone.

It's to wow old people with your new Dick Tracy watch.

I'm only partly joking.

 

[lopri]

To act like a hidden cam? lol. I am not a watch person so this product isn't for me, but the idea of having to charge your watch every day is quite unattractive.

 

[Crono]

To act like a hidden cam? lol. I am not a watch person so this product isn't for me, but the idea of having to charge your watch every day is quite unattractive.

Yeah, that's what makes this more a novelty than a useable device.

If they could up the battery life and make a flexible wrist computer - a long band more than a "watch" - I'd buy it. Something like what the Predators had (but more advanced), or featured in shows like Chuck or Continuum.

You would think Samsung with all their OLED tech would try and jump straight ahead to this, but I guess the technology isn't there yet. I think Apple might make something in between the Galaxy Gear and what I want, but with more of a Fuel band look and featureset.

Instead, what we are getting is this:

It's the same way in the larger mobile device space. What I really want are flexible scrolls, but instead we get tablets first.

 

[kyrax12]

Isn't amoled screen hard to see in Sunlight. LIke incredibly hard?

 

[lopri]

It's due to low brightness. Under a different sunlight it's actually better because it's less reflective. (I don't know if it's a characteristic of AMOLED or the glass but it looked that way to me, comparing the S4 and the N4)

 

[joshhedge]

What's the point of a watch which doesn't display the time, all the time? I think Samsung have made a poor implementation of a 'smart watch' right here. I'd like to see what Apple brings to the table, although I'm not going to buy either.

Having to charge it every day is a shocking design flaw. Granted battery technology has to vastly improve in the mobile space, but in my opinion that's not acceptable for a watch of any kind. The camera seems like a lovely gimmick as well. Looking positively, it is a first gen product, so it can only be improved.

 

[slashbinslashbash]

Yeah, this seems half-baked at best. Glad to get Anand's reporting on it. The camera is positioned horribly. I would much rather see it on the main part of the device itself.

1) The SoC is way too fast. 800MHz! Big-time power suck. Should be cut to 400Mhz if not to 200Mhz or less. Leave the heavy computing to the connected smartphone with the big battery. The display is 320x320, that is 3/4 of the original iPhone's resolution. Shouldn't need a tremendous amount of compute or GPU power. With appropriate communications, you could even have rendering done on the smartphone and the output sent to the watch. Think big.LITTLE, except the watch only has the LITTLE, and any "big" stuff comes from the phone.

2) However, of course the main power problem is not the SoC but rather the display. I'm envisioning something like maybe a hybrid display, with a transparent e-Ink panel stacked on top of a color LCD? So in "passive" mode the e-Ink is on, showing the time and any alerts/messages that come up, without having to fire up the backlight. Once the watch is "activated" then the top display goes transparent, allowing the color LCD to shine through.

I actually like the Pebble idea quite a bit. Given current tech, I think Pebble is a pretty good compromise between battery life and usefulness. I would easily take a Pebble over this Samsung (assuming I had a Samsung phone to connect it to). However, I don't expect Pebble to catch on outside of the techie crowd and maybe a few other niches like fitness people. Apple would need to juice it up considerably in order to attract a mass audience. I think that a color screen is necessary to attract that audience; as well as some kind of "always on" display.

I also think that water resistance is necessary. Hate to ruin a $300 piece of kit by washing your hands or getting caught in the rain. They need to either eliminate all ports, or have them gasketed. I haven't seen anything that mentions this for the Samsung, so I guess it's not water resistant?

 

[joshhedge]

Yeah, this seems half-baked at best. Glad to get Anand's reporting on it. The camera is positioned horribly. I would much rather see it on the main part of the device itself.

1) The SoC is way too fast. 800MHz! Big-time power suck. Should be cut to 400Mhz if not to 200Mhz or less. Leave the heavy computing to the connected smartphone with the big battery. The display is 320x320, that is 3/4 of the original iPhone's resolution. Shouldn't need a tremendous amount of compute or GPU power. With appropriate communications, you could even have rendering done on the smartphone and the output sent to the watch. Think big.LITTLE, except the watch only has the LITTLE, and any "big" stuff comes from the phone.

2) However, of course the main power problem is not the SoC but rather the display. I'm envisioning something like maybe a hybrid display, with a transparent e-Ink panel stacked on top of a color LCD? So in "passive" mode the e-Ink is on, showing the time and any alerts/messages that come up, without having to fire up the backlight. Once the watch is "activated" then the top display goes transparent, allowing the color LCD to shine through.

I actually like the Pebble idea quite a bit. Given current tech, I think Pebble is a pretty good compromise between battery life and usefulness. I would easily take a Pebble over this Samsung (assuming I had a Samsung phone to connect it to). However, I don't expect Pebble to catch on outside of the techie crowd and maybe a few other niches like fitness people. Apple would need to juice it up considerably in order to attract a mass audience. I think that a color screen is necessary to attract that audience; as well as some kind of "always on" display.

I also think that water resistance is necessary. Hate to ruin a $300 piece of kit by washing your hands or getting caught in the rain. They need to either eliminate all ports, or have them gasketed. I haven't seen anything that mentions this for the Samsung, so I guess it's not water resistant?

Regarding point 1, it's running on Android, apparently. So a slow SoC is going to be an absolute dog.

 

[Dari]

Relax everyone. Once Apple makes their watch, Samsung will ramp up the copying machine and all will be well. In the mean time you can buy smart watches from SONY and Pebble that actually last days...

 

[slashbinslashbash]

Regarding point 1, it's running on Android, apparently. So a slow SoC is going to be an absolute dog.

But "Android" isn't a single monolithic OS. They can strip out the pieces they don't need, such as most of the network stack, arbitrary app support, most 3D rendering support, etc.... if it is not a 90% stripped down version, then Samsung is just being stupid. Nobody is going to expect to play full-on games on this tiny watch screen.

It is perfectly reasonable to restrict what apps are installable on this device. In fact, apps will need to be (re)written specifically for this device, if for no other reason than its tiny size. Precise pointing is not going to happen; the control method will have to rely more on swipes and gestures than touches. I have a 6G iPod Nano, which has a 1.5" screen size, and the best precision you can really rely on is dividing up the screen into quadrants. The rest is swipes. Anything else, your fingers are simply too big and/or your hand blocks your view of the screen. This device should not be able to run any arbitrary Android app, period.

 

[TheStu]

But "Android" isn't a single monolithic OS. They can strip out the pieces they don't need, such as most of the network stack, arbitrary app support, most 3D rendering support, etc.... if it is not a 90% stripped down version, then Samsung is just being stupid. Nobody is going to expect to play full-on games on this tiny watch screen.

It is perfectly reasonable to restrict what apps are installable on this device. In fact, apps will need to be (re)written specifically for this device, if for no other reason than its tiny size. Precise pointing is not going to happen; the control method will have to rely more on swipes and gestures than touches. I have a 6G iPod Nano, which has a 1.5" screen size, and the best precision you can really rely on is dividing up the screen into quadrants. The rest is swipes. Anything else, your fingers are simply too big and/or your hand blocks your view of the screen. This device should not be able to run any arbitrary Android app, period.

Yea, but doesn't that require Samsung to show that they have some software chops? I'm pretty naive on Samsung's software, other than the few times that I have messed around with my mom's GS4, but I didn't exactly come away impressed by what Samsung had brought to the table.

 

[nOOky]

It could be just me, but I don't see the usefulness of a watch that at best mimics a smartphone, especially when it sucks as a watch. Why not just pull your phone out of your pocket and take a picture or look at the time if you need to turn the watch on anyway?

A better use of a watch is as a fitness GPS device imho. The Motorola Motoactv was ahead of it's time in this regard, except for the water resistance rating it was a nifty device. I don't carry my phone with me when I run so a watch like the Samsung, even with Myfitnesspal is useless unless you carry your phone.

Gimme something useful, like a water proof GPS watch with a built-in mp3 player that I can mount on my wrist or upper arm, not an ugly watch that sucks as a watch and gets crappy battery life. I don't want to watch videos on a watch, or look at pictures, or even view texts, I just want to see the time when I look down at it.

Of course I'm not surprised by the Samsung watch not making a very good watch, most smartphones don't do a good job of functioning as a phone either.

 

[Joe1987]

I'm beginning to see why Apple hasn't released a watch yet, the Galaxy Gear isn't something I'm willing to wear. Maybe in a couple of generations.

 

[cbrunny]

Won't be buying one of these things any time soon - at least not until they wake up and make the thing round like a normal person watch. Seems so obvious to me... Avoid the blocky look by avoiding the blocks.

 

[TheStu]

Won't be buying one of these things any time soon - at least not until they wake up and make the thing round like a normal person watch. Seems so obvious to me... Avoid the blocky look by avoiding the blocks.

As soon as they can effectively make round LCDs that allow for all kinds of usage paradigms...

 

[Red Storm]

It should have been e-ink, ditch the camera and keep the calling capabilities.

 

[ControlD]

Yea, but doesn't that require Samsung to show that they have some software chops? I'm pretty naive on Samsung's software, other than the few times that I have messed around with my mom's GS4, but I didn't exactly come away impressed by what Samsung had brought to the table.

My experience with the Note II has been quite the opposite actually. I am fairly impressed with all that Samsung added to that device.

 

[bearxor]

It should have been e-ink, ditch the camera and keep the calling capabilities.

I don't think it should have been e-ink if you want it to act like a regular watch. e-ink isn't really that power efficient if you're constantly refreshing the screen, which is what a watch would be doing to count down the seconds. It would also be hard to incorporate certain features, like a stop watch, because of the poor refresh rates of e-ink.

Pentile AMOLED is made for stuff like this. It's practically perfect. Have an always on display showing the time with only the pixels you need on.

What this thing really needed is a way slower CPU and a slightly bigger battery. It doesn't need a week of power, but three days would have been sufficient.

 

[MrX8503]

I think it'll be awhile until we get a knockout hit with a smart watch. A smart watch has to be able to do tasks better than a smartphone or tablet, otherwise it's junk. On top of that, they need to be tasks that people find a necessity in their every day lives. What these tasks are and how they'll be executed is a mystery, but that's what will make it a knockout.

Isn't amoled screen hard to see in Sunlight. LIke incredibly hard?

Yes, so OLED is probably not a good candidate for a watch.

It's due to low brightness. Under a different sunlight it's actually better because it's less reflective. (I don't know if it's a characteristic of AMOLED or the glass but it looked that way to me, comparing the S4 and the N4)

The brightness is low because if it were to have the same brightness as a conventional LCD the battery life would be destroyed. This is why you can't manually max out the brightness on an S4.

The S4 has the least reflectance glass and really doesn't have anything to do with OLED (correct me if I'm wrong on the OLED part). Samsung was smart and did this so that the S4 was more viewable in direct sunlight.

 

[dguy6789]

Any smartwatch that has the screen behave like a smartphone is an automatic deal breaker for me. It has to display the time at all times without requiring me to do any kind of gestures or press any buttons. The point of a watch is to get the time at a glance. A smartwatch absolutely can't be worse at being a watch than what it is replacing.

 

[Eug]

Wow, this thing is totally pointless.
Ditto for Sony's.

Quite honestly, if I were forced to wear a watch, I'd rather have this:

Or even better, this:

 

[cbrunny]

As soon as they can effectively make round LCDs that allow for all kinds of usage paradigms...

Not sure I understand. Round LCD's are not possible?

http://www.electronics-eetimes.com/...als-or-clocks.html?cmp_id=7&news_id=222901229

http://gizmodo.com/312126/toshiba-matsushita-develops-round-lcd-display

http://spectrum.ieee.org/computing/hardware/circular-lcd-debuts

 

[ControlD]

Gimme something useful, like a water proof GPS watch with a built-in mp3 player that I can mount on my wrist or upper arm, not an ugly watch that sucks as a watch and gets crappy battery life. I don't want to watch videos on a watch, or look at pictures, or even view texts, I just want to see the time when I look down at it.

Agreed. I would buy a device like that for sure. I think that is sort of what Apple was going for with the previous version of the Nano, but it ended up being a terrible POS.