It's a 10 year old computer that can take full advantage of my gigabit home internet connection (940Mbps up and down) - something only dreamed about 10 years ago - can run the latest OS and can be upgraded to have the same amount of RAM as most computers that are sold today.
In 2008, a 10 year old computer would have probably had these specs and be unusable as a modern computer.
http://www.tomshardware.com/forum/266641-28-average-1998-com...
To sum it up, a decent compy around 1998/early 1999 could look likte this:
400Mhz P2 64mb sdram running at 100 mhz one or two voodoo 2 cards (90mhz core, 8 or 12 mb EDO VRAM) a 2D card (voodoo 2 was a 3d accelerator only) ATA-2 HD
In 1998 a 10 year old computer would have been less usable.
How fast improvements come to the smart watch will slow down in time but it's a brand new product.
o Your Plex system is a home server, which have always typically been lower powered desktops even 10 year ago (I know because I was building home servers back then. In fact 10 years ago I was running ZFS off an old desktop PC with some really budget consumer RAID controllers)
o and you're comparing that to a gaming machine which are systems which are typically at the higher end of spectrum). To put things into perspective, in 1999 I most certainly did not have a 3D accelerator. Yet I was still using my computer for software development and web development (including, by the way, 3D markup via VRML). In fact I don't think I even had a P2 considering my upgrade a few years later was to Celeron. I have a feeling my '98 PC was a Cyrix.
o It was really common even in the 90s to run old PCs. Even in 1995 I was still seeing 386s and even the occasional 286 in use running Windows 3.x.
o Lastly you cannot really compare a watch to a PC. Sure, smart watches are basically just computers these days; but the point the OP was making was peoples expectations are different for watches. I mean we have digital watches from the 80s that still work as well today as they ever have done, yet smart watches made 2 years ago are already junk. Before people comment on the difference in tech, yes I already know. But that's not the point. The point is people's expectations - which frankly I think are more than reasonable. If there's anything we have learned in the last 20 years it's how to futureproof our products.
The Plex Server has to transcode all types of formats to H.264 in real time. My old 2006 Core Duo 1.66Ghz Mac couldn't handle it. It is still being used by my parents running Windows 7. I also tried using a 2Ghz Pentium Dual Core.
* (I know because I was building home servers back then. In fact 10 years ago I was running ZFS off an old desktop PC with some really budget consumer RAID controllers)*
They weren't transcoding 1080 video to H.264 real time in 2008.
o and you're comparing that to a gaming machine which are systems which are typically at the higher end of spectrum). To put things into perspective, in 1999 I most certainly did not have a 3D accelerator. Yet I was still using my computer for software development and web development (including, by the way, 3D markup via VRML). In fact I don't think I even had a P2 considering my upgrade a few years later was to Celeron. I have a feeling my '98 PC was a Cyrix.
Would you still be using that computer in 2008? Could it run the latest version of Windows? Could it take advantage of the fastest speed internet that was available to home users in 2008?
o It was really common even in the 90s to run old PCs. Even in 1995 I was still seeing 386s and even the occasional 286 in use running Windows 3.x.
My 2008 Core 2 Duo can run the latest software and the latest version of Windows well. Could those 10 year old computers run Windows 98 in 1998?
Lastly you cannot really compare a watch to a PC. Sure, smart watches are basically just computers these days; but the point the OP was making was peoples expectations are different for watches.
How is a watch that runs a phone OS that makes phone calls not considered more of a phone than a watch?
I mean we have digital watches from the 80s that still work as well today as they ever have done,
The difference is that my 10 year old computer can do the same thing as a modern computer - run the latest OS and software, has just as much memory and storage (HDD vs SsD) and has a slightly better display (1920x1200 vs 1930x1080). Can that 10 year old digital watch do the same as a modern smart watch?
If there's anything we have learned in the last 20 years it's how to futureproof our products.
The only reason that my old Core 2 Duo is "future proofed" is because useful improvements have slowed down dramatically in the last 10 years. In 1998 the typical computer was coming with 32-64Gb of RAM and couldn't be upgraded to 4GB. My computer from 2008 came with 4GB RAM, the same amount as some low end computers today and can be upgraded to 8GB.
My 2008 computer has gigabit Ethernet. Which is still the fastest consumer internet connection.
I have four Core2Duo tower computers that I bought last year for $10 each. Three of them have 8GB of RAM and the other has 4GB. Each one has an Nvidia 1050 or 1060 GPU in it and they run TensorFlow just fine, though the computers are PCIe 2.0, so I'm "only" getting a little over 5 GB/s data transfer between the cards and the CPU. Gigabit ethernet makes that mostly irrelevant.
For the most part, when the GPUs are running all-out on something, the CPU is 75% idle or more.
That that's more than 10 years old. Your original point was that you're using a 10 year old system.
> They weren't transcoding 1080 video to H.264 real time in 2008.
Again you're comparing apples to oranges because most people wouldn't have had a need for transcoding 1080p videos from home servers back then. Most broadcasters still weren't sending HD streams or had only just started to. So a 10 year old PC in 2008 doing HD would be like a 10 year old PC in 2018 doing 4k.
>* My 2008 Core 2 Duo can run the latest software and the latest version of Windows well. Could those 10 year old computers run Windows 98 in 1998?*
The examples I gave were from around 1994/5 and as I had already cited they were running Windows 3.x (Windows 95 hadn't been released at that time). So yes, the latest OS.
> How is a watch that runs a phone OS that makes phone calls not considered more of a phone than a watch?
I'd already address that point as well. One word: "expectations". I don't expect a watch to be obsoleted after 2 years regardless of the tech it includes. If said tech means my expectations cannot be met then I simply will not buy any more watches made by that company. After all, I already have a phone; what I need is a watch.
> The difference is that my 10 year old computer can do the same thing as a modern computer - run the latest OS and software, has just as much memory and storage (HDD vs SsD) and has a slightly better display (1920x1200 vs 1930x1080). Can that 10 year old digital watch do the same as a modern smart watch?
I've already addressed that point too. Watches are not PCs (forget the technology and thing about it's form factor and what it's actual usage is). Thus comparing the two doesn't make much sense.
>* The only reason that my old Core 2 Duo is "future proofed" is because useful improvements have slowed down dramatically in the last 10 years. In 1998 the typical computer was coming with 32-64Gb of RAM and couldn't be upgraded to 4GB. My computer from 2008 came with 4GB RAM, the same amount as some low end computers today and can be upgraded to 8GB.*
Nope. I've already demonstrated that software likes to keep up with hardware. This it as true now as it was 10 years ago (even 20). What's changed is:
o we've moved back towards a time sharing model where processing is offloaded to the cloud
o we've learned lessons about security after the mistakes of the 90s and 00s where PCs were basically just honeypots for malware. Granted we still make lots of mistakes even now (particularly around the area of IoT where we seemed doomed to repeat past mistakes) but in terms of actual PCs the groundwork was put in 10 years ago.
o software development processes have matured and also include better automated testing frameworks making it easier to detect problems and test on a wide range of platforms
o Microsoft also got their fingers burnt with the whole "Vista Compatible" thing on backstock Intels that clearly weren't capable. So invested more in the future releases of Windows to ensure that didn't happen again.
You could also argue that the pace of "useful improvements" in OS development has sped up over the last 10 years because more time has been invested in making platforms stable, secure and performant instead of just throwing new shiney feature after new shiney feature.
>* My 2008 computer has gigabit Ethernet. Which is still the fastest consumer internet connection.*
Another meaningless comparison because my 1998 computer had 100Mb ethernet, which was several orders of magnitude faster than the then fastest consumer internet connection. If anything I think the state of networking gear has stagnated because I was handling gigabit switches back in 1998 (with AppleTalk - yuk), which took forever to reach consumers and still many ISP routers only offer 1 complimentary gigabit port. Yet most offices / server rooms are still only on gigabit switches some 20 years later and most home users are probably on WiFi or Powerline adapters (neither of which will come even close to 1GbE). So I guess the real question is, how fast does one actually need their LAN to be? Either way, it's wildly off topic.
My 10 year old computer is doing real time transcoding in 2018.
The examples I gave were from around 1994/5 and as I had already cited they were running Windows 3.x (Windows 95 hadn't been released at that time). So yes, the latest OS.
Again. That's the point. My 10 year old computer is running the current latest OS in 2018. A 10 year old computer in 2008 couldn't run the latest OS.
My major point being that in the first few facades of PCs, technology was moving so fast that to run the latest software, you had to upgrade your computer more frequently. Now, a 10 year old computer can run the latest software.
The smart watch and the phone to a lesser extent is still seeing the type of rapid improvements that you saw in computers until 10 years ago.
An electronic watch from 10 years ago can't do everything a modern smart watch can. A 10 year old computer can.
As far as the actual usage. A modern smart watch is used for a lot of things that a 10 year old watch wasn't.
Yes, I got that point. And I'm saying 10 year old computers in 2008 were doing transcoding as well - just with standard definition streams rather than HD (the "HD" part seemed important to you but I was making the point that you're not doing 4k transcoding in 2018 with your 10 year old PC so why fixate on HD recording in 2008 when HD was just as new as 4k is today)
>* Again. That's the point. My 10 year old computer is running the current latest OS in 2018. A 10 year old computer in 2008 couldn't run the latest OS.*
Maybe I'm not making myself clear enough because this is the 3rd time I've now had to say that I've used 10 year old computers both in 2008 and in 1998 - all of which were running the latest OS. Like I said, in 1995 I was still supporting Windows 3.x on a 286s.
> My major point being that in the first few facades of PCs, technology was moving so fast that to run the latest software, you had to upgrade your computer more frequently. Now, a 10 year old computer can run the latest software.
And I'm saying that is bullshit. I have first hand experience - originally as an owner and a support engineer; and then as a developer and systems administrator - with examples over the last 20 years that directly disprove your point. Examples I've repeatedly cited. Frankly if I've seen any trend it's people replacing hardware quicker now that laptops have gotten so cheap - which is the exact opposite of the point you're trying to make.
> The smart watch and the phone to a lesser extent is still seeing the type of rapid improvements that you saw in computers until 10 years ago.
I've repeatedly demonstrated that your argument about rapid developments in the PC landscape is completely baseless and given numerous examples why.
>An electronic watch from 10 years ago can't do everything a modern smart watch can. A 10 year old computer can.
Again you're comparing apples to oranges. A watch is not a PC. Hell why don't we just compare it to cars and have the same meaningless discussion: a digital watch is like a small hatchback and a smartwatch is like a campervan. What we've now proved is that smartwatches have running water....oh wait.
> As far as the actual usage. A modern smart watch is used for a lot of things that a 10 year old watch wasn't.
That is literally the only accurate thing you've posted. The real question is whether the other stuff is enough to make people want to upgrade their watches (as well as their phones) every two years. I don't think it is. I think those kind of people are the exception and most people wouldn't want their smartwatch to become obsolete so quickly. I think most people see their smartwatch as a digital watch on steroids rather than a PC. And thus I think most people's expectations would be for their watch to outlast their phone.
At risk of repeating myself (but let's be honest, you've missed the other points I made on 2 separate occasions) the hardware is irrelevant to most people. Most people aren't techy. Most people will look at a smartwatch and see "something-watch" not "computer-something" like you and I.
Turned out that it runs OS X El Capitan (= gets security updates) and supports 8 GB of RAM. You can upgrade the 2.5" SATA HDD to SSD. Basically it's usable as a modern computer, only the 1280x800 display resolution being modest by today's standards.
In fact, I'll probably be fine if third-party apps stop working on the series 0 entirely.
Third-party apps run so poorly on the series 0 that they're not really worth using. I'd rather developers invest in the work that will make later-generation Apple Watches worth the upgrade, than pour those resources into low-value results on a hamstrung platform. My current watch won't be any worse off than the day I bought it, and I'll have the option of a more compelling upgrade if I want one.
As a consumer, I don't expect a first-generation smartwatch, especially from 2015 when this was still a very new product category, to last 5-10 years. If it does last, I don't expect it to acquire new features every year. Other watches don't. Appliances that last that long tend to get worse over their lifetime.
I may be an outlier here because I came to the Apple Watch from the world of higher-end watches. My usual daily wear piece before was an Omega, e.g.
I started wearing the Apple Watch when I was training for a half marathon, because it worked really well for that purpose (when paired with Strava) -- and I discovered it was also so USEFUL for me that I wore it almost all the time.
After about 18 months, I realized I was really only taking it off when I felt I needed to wear something that looked a bit fancier -- but I didn't want to give up the Apple functionality. Then the series 3 was introduced, with actual water-resistance and the in-built cell capability, and that was that. I sold my initial one (a Sport model) and bought the next nicer model in steel with the sapphire crystal on the Milanese loop. I swap in a composite band if I'm being active, and wear the loop if I'm going to dinner.
I don't expect this watch -- about $750US -- to last 10 years, but I will be happy if I get 3 out of it before I want to upgrade. I also expect it'll still WORK at that point, and the driver for me will be new features or functions and not failure. I'm okay with that, because of the relatively modest cost vs. definitively high usability/duty cycle.
If I were the sort of person who would wear a sport model all the time, it'd be an even easier choice ($359 w/o cell; $429 with).
Wearable tech has already been around for more than 10 years. Whether it's raincoats with speakers built into the collar, or the digital watches from the 80s. The problem isn't so much wearable tech but rather current tech trends (eg the IoT revolution). To that end, I'd wager something like Pebble (baring hardware failure) would still work in 10 years since the only dependency that has is bluetooth.
I doubt it. The watch will still work, but the app needed to sync to the watch won’t.
I’ve been really frustrated with old “utility” apps not working anymore on iOS (due to not being updated to 64bit). I’m all for upgrades and new stuff, but it sucks when a simple app that did exactly what you needed is no longer able to run.
I think it's really about the cloud. I kept my last iPhone way past a typical phone's lifespan, and got to experience watching it slowly revert back to being a dumbphone. In the same way that smart home devices have a nasty tendency to stop working as soon as the vendor decides to stop maintaining some critical server the product needs to be able to ping just to turn on properly, apps eventually stop working with the latest OS your phone can run, and pretty shortly after that some service running in the cloud will stop being able to talk to whatever version of the app you're stuck on.
In the case of Apple's or AT&T's apps, this happened almost immediately after the first version of iOS I couldn't upgrade to came down the pipe. In the case of others, it was 6 months or a year later. The web browser died in its own special way, choked to death by more JavaScript than it could handle.
I get the economics behind why companies do this, but it's sad, all the same. A 35 year old Commodore 64 is every bit as fun as it was when it first came out, and every bit as capable, but a piece of tech that's only 3.5 years old probably has a blinking jewel in the palm of its hand.
I think you have odd perspective here, because you're talking about tech lasting for a decade, but you're also explicitly omitting anything with a computer in it ("run an operating system").
"A 35 year old Commodore 64 is every bit as fun as it was when it first came out, and every bit as capable, but a piece of tech that's only 3.5 years old probably has a blinking jewel in the palm of its hand."
The hypothetical 35-year-old Commodore, however, can't do most of the things I ask my 3 year old laptop to do. Durability wasn't the issue; the rapidly increasing capability of newer and more capable computing platforms was.
We're at a point now where Moore's Law & related phenomena seem to have slowed a bit. Ten or fifteen years ago I always upgraded my laptop every 3 years. Now I can go longer, and typically only upgrade when there's a compelling reason (higher RAM ceiling, e.g., or external factors like the economics of extended service plans). Phones are the same; in the earlier years of the cell phone era, I got a new phone nearly every year because that meant smaller and more powerful and therefore more useful. In the Smartphone era, two years became the norm, and nowadays it's routine to see folks with 3 or even 4 year old phones.
The tl;dr is that there's no conspiracy or malignant intent here; it's just that shit gets better, and so the upgrade makes sense for most people. I mean, maybe you can get a 35-year-old Commodore running as a novelty, but I sure wouldn't want to try to do my job with one.
No, but it can do all the things it promised to do when it was first sold.
By contrast, one of my old iPads is thinking about enrolling in a job retraining program to learn to be a trivet. This despite all of its components being in perfect working order.
I never suggested this was a conspiracy, just that it's a travesty.
FWIW I think developers should support old hardware as long as they can, but only the lastest operating system for each particular supported model.
To some extent, RMS is right on this subject: When your devices rely on the cloud, it means you don't really own them, because any cloud-dependent functionality continues to exist at the pleasure of others.
Yes, and no.
I have a launch-day iPad, and it became less and less useful over the years as the apps stopped being supported by their developers.
The last version of Safari for the original iPad is unusable if you surf anywhere but HN. It might as well be Lynx.
This time last year, my iPad was down to just e-mail, notes, streaming video from my satellite box, PBS video, and a great app from Panic called Status Board.
Now it's down to just e-mail and notes, so it spends more time in the desk drawer than propped up on a tiny easel on top of the desk.
I have an iPad Air 1, a device that is pushing ~5 years by now, that became abhorrently laggy with the transition from iOS 10 to iOS 11. Sliding up the multitask view or sliding down the cover sheet usually hovered around 5-20FPS, and sometimes came with a 1 second (!) delay. I don't know what magic the iOS devs worked with the upcoming iOS 11.3 (I run the beta) but both actions are fluid now, and the delay is gone. Besides that, my parents have an iPad 4, which is 'stuck' on iOS 10 but still is incredibly smooth and, so far, works perfectly well with iCloud.
I used to think that Apple didn't mind old devices being bogged down with big updates, since the unintended side effect would be people pushed to newer devices. However, with techniques like App Thinning, the way the 5S became smooth again going from iOS 9 to 10, and now the way they revived my iPad Air I'd say they seem to be reasonable in choosing when devices are too old and holding back iOS as a whole. Hell, with Apple your iOS device gets 4-6 years worth of big updates.. with Android, you're lucky if you get 2.
This applies even more for watches.
For me the term seems to be one of those which got twisted by marketing types to mean something wholly different to encourage this buy the new model mentality.
But I don't use it! It gets no security patches.
If I wanted it for alarms, notes, etc it would be fine
I also don't completely understand the tone, this dev is measuring the installed base of different models. If their users are suddenly < 5% on the seminal model of the watch, then screw halting progress for 95% of your users just to hold on to that 5%. Expectations shouldn't be measured in years, it should be measured in adoption base, and optimizing for where your users are. And that number is going to wildly swing based on maturity of the tech, and that's going to change over time, so trying to throw out a year range as table stakes to apply forever is just a little crazy.
The Apple Watch is only a watch incidentally, all indications seem to be that it's mostly a fitness tracker with a bit of notification monitoring thrown in. Yes, it does happen to tell time as well. I expect there to significant enhancements to the fitness/health tracking capabilities in the next 2-3 years (and then the 2-3 after that as well), some of us are fine with upgrading a fitness tracker every few years.
As an indie dev* making a living from Apple Watch, I concur with the difficulty of having to debug these things.
For example, some arbitrary resource limit may have transparently been applied, leading to some unexpected behaviour in your app.
Now, make a minor change to code to diagnose this issue and deploy to device - could take 2-3 minutes to copy to device, and another 30 seconds to run the binary.
Series 3 still has its issues but it is 1000x better than the original.
* https://outcastapp.com, plus various others
I expect my Rolex to last a hundred years and my Apple Watch to last as long as my phone. Especially since technically that is what it largely is. A small phone on your wrist.
I'll never understand the consumer dynamic that allows there to be a never-ending lineup of $800+ phones at every store that sells electronics. Congratulations, your [electronic device] was expensive and in about two years, you're going to need a new one, just the same.
Once I got a smartphone with a data plan, it turned out to have a fairly high value in my life. $2/day roughly covers my data plan and buying a new $800 phone every two years. The consumer dynamic is that having the latest phone is worth $2/day.
Dropping $800 recently on a new phone did make me feel something, but it was a positive feeling.
I wouldn't expect new features but I would expect vulnerabilities to be patched and any cloud dependencies to remain switched on.
But it also ONLY tells the time and the date, and the date will be wrong if you don't stay on top of it because it doesn't know which months have 31 days and which are shorter.
It's super cool, though, that this object still works just as well as when it was new in 1978.
But I stopped wearing it (and my other mechanicals) all the time precisely because of their limited featureset vs. a good and attractive smartwatch.
Non Smart Watches stop getting updates the moment they are manufactured. Apple Watches getting updates for 3 years, is good enough.
I’m saying this as someone who uses Watch Series 0 everyday.
Any more updates to this old watch and it’ll only get slower.
From a developer perspective also, there’s no point supporting such an old device when it wasn’t built for anything more than the current functionality.
For the price of a Rolex or an Omega, you can ‘probably’ buy a new Apple Watch every 3 years, for the rest of your life. It’s a personal choice.
Your last sentence isn’t often correct either: I have a 2012 Toshiba TV. The hardware is in great shape but of all of the apps it shipped with only the Netflix and Vudu apps still work because everything else broke when Google, Yahoo!, etc. stopped SHA-1 support. I agree that no new features is a reasonable policy but at a minimum you need basic security & compatibility support for 5-10 years.
1. We’re talking about hardware vendors 2. Why is the tech industry magically different from everything else? I get safety recalls on things which are 5-10 years old but e.g. Lenovo can’t seem to ship Android security updates which Google gave them more than 1-2 years after release.
> Sadly, on Apple platforms there is no way of doing paid upgrades so there goes that.
This is why developers use advertising, in-app purchases and subscriptions, or releasing major versions as new apps.
The point wasn’t that the status quo is perfect but that it’s extremely wasteful and needs to change. The tech industry has a lot of people who thought that cycle of rapid growth was innate but are starting to realize that things are saturating in most segments. Buying a new device every 2 years only works if they’re very cheap or dramatically better.
From the standpoint of hardware I definitely agree that it should not be made with a 2-3 year lifespan in mind, especially because of the toll it takes on the environment.
> This is why developers use advertising, in-app purchases and subscriptions, or releasing major versions as new apps.
None of these are a good replacement because they do not solve the problem. If you wish to have an old version of the app available for users which did not update the only option is releasing a new app but then you can not give a discount to old clients. In app purchases and subscriptions pull everybody into the release cycle.
The current status quo is not good but from the software perspective I do not know how it could change. The prices were pulled down so far that supporting old platforms is not sustainable, unless you are one of those companies which get their money somewhere else.
There are gadget lovers who get the latest at all ages, and tons of them are at the 'dad' age (30-40+), not younger.