Universal Mac app for new M1 Apple Silicon

Eventually, more than likely, electron will have to update for better iMac compatibility. Otherwise they will go the way of java applets, gwt, java swing and swing2, and so many other programming languages that have tries to please everyone and end up pleasing no one.

My point here is that TR is doing nothing wrong. And it’s not their fault that software they use to create their desktop apps suddenly suck.

We are asking TR to switch to a new code base depending on the platform. That mean developing teams for both Intel chips, and arm based chips. This means that all apps will not be in sync, they might looks different, they might have different behaviors and overall might not give similar experiences to all users.

As a software developer i can tell you all this is a very expensive ask. Software developer are VERY expensive now days. TR would need to hire more engineers to support this in a way that can mitigate all the risk i mentioned above.

All of this in the same breath that we ask TR to not take the legacy pricing away and that not rise the price even further, while also asking them for a billion different features.

Perspective people. Non of what you are asking is free nor we want to pay the price for it.

Does it sucks. Yes, it does. But I don’t think blaming TR for this is necessarily fair to them…

3 Likes

I don’t think so. Usually, the point of cross-platform development tools is to develop for the lowest common denominator. Most apps are cross-platform, and it shows. Just think of banking apps, Facebook, Google Maps, etc. The only sort-of exception are games, but here typically the game engine gets ported by the game engine developer and the game sits on top. However, standout apps on iOS, macOS and iPadOS are much more likely to be written with native APIs (e. g. Omni’s apps or Pixelmator come to mind).

But there is a strategy tax and IMHO it is mounting. Right now, TR is missing tons of health data (sleep, HRV, resting heart rate, etc.), because it is not plugged into Apple’s Health Kit. There is no Apple Watch integration either (again, less data). And they need to develop around limitations and think about re-implementing platform-specific features anyway. I don’t think it is possible to replace these, because all alternatives are worse.

For example, the problem with split view that @rac1ngsnak3 posted could likely be addressed more easily and systematically with Apple’s iOS/iPadOS size classes. Because size classes overlap, I reckon it’d lighten the workload in more ways than one. (Although to be fair, I don’t know how Electron handles this, probably by inquiring the screen size just like websites would?)

Importantly, the companies who do cross platform development most successfully (e. g. Adobe or Microsoft) in the end rely crucially on their own cross-platform frameworks. And once you are there, I really don’t know whether this is less effort. In my understanding, it is about consistency (e. g. being able to open documents and have them render in exactly the same way).

Fair enough. But we are all paying for it. I’d be willing to pay 2x what I pay now if they released native clients with the added functionality. No joke.

Or me. Got my first Mac on what would become ebay Germany in 1997. (An impulse buy, I wanted a notebook at the time.)

What is worse, I worked for Apple for a bit, checking their localization, which gave me OCD from hell when it comes to UI inconsistencies of all stripes.

1 Like

I wonder what the device composition is. Like what is the percentage of people on Mac vs PC vs Android vs iOS? @Nate_Pearson any insight?

1 Like

You are comparing Apples (pun indented) to oranges. This are multi billion dollar companies with resources on each team to develop apps for individual OS. There is a reason most of the apps looks different and have different setting on each OS. TR doesnt have that kind of capabilities (read people/money) right now. They are probably more focused on backend capabilities than front end (think about it… when was the last time the WC was updated?)

Are we paying for the front end or for ALL the other stuff? I know we are paying for all of it. And I think we are getting a lot of value out of what we pay with all the backend updates.
You would pay more. I would probably pay more. But if the legacy price thread is an indication, there are MANY who want to pay the least amount and get everything forever.

2 Likes

I don’t recall at what point TR released their app on iOS but ever since there are probably just a couple times I’ve run the app on anything other than an iPhone or iPad. And 99.999% of the time I just run it on my iPhone. Obviously from my perspective I don’t see the great need for TR to develop native environment versions of their app unless it really is causing big problems or they are getting a ton of complaints.

1 Like

This is on latest macOS:

2 Likes

Good info here. Looks like no Electron in the iPhone/iPad app at least. However, got some baffling news that TR doesn’t support iPadOS multitasking, which is one of the best/biggest features of iPadOS 15.

From TR support:

Split-screen usage is not an officially supported feature at the moment, which means it is not part of our stringent testing protocol that we undergo each time we release updates to the app. That being said, in initial testing, it does seem to work and many of our athletes are able to train while multitasking successfully using this feature.

For future use, we try to recommend not using split view since there will be some interface or connectivity issues as you are definitely experiencing.

It boggles my mind that after 5 months since release of iPadOS 15 and close to a year since WWDC where they announced split-view, that this fundamental feature isn’t supported. Instead they’ve been working on AI FTP, which is worthless and redundant.

1 Like

:laughing: different perspectives I guess … given a straight up choice, I’d take AI FTP over split screen support any day … and I’d be surprised if I’m the only one.

9 Likes

based on feedback on BOTH things (the lack of proper Mac binaries vs AI FTP)… I think the vast majority (or at least on the forum) would disagree with this statement.

5 Likes

I’ve been reading this thread since it started and I still don’t get the urgency some of you have for a proper Mac binary. The current app works fine. Seriously, we’re looking at blue bars! Aside from doing group workouts, TR is not some resource hog. Furthermore, for the new M1 and M2 Macs, most if not all apps run faster in Rosetta emulation than on the old Intel Macs.

Yes, I’m a Macbook Pro (Intel) user. In fact, I’m about to upgrade to the new 14in MBP.

Am I missing something?

edit: “…proper new M1/M2 binary…”

2 Likes

There are something that doesn’t work well on the non Intel Macs.

I have no idea what they are since I use an ipad…but seems enough to frustrate a lot of people.

Split screen is older than that, and many cross-platform apps struggle with that.

1 Like

I agree, TR can’t follow the same strategy here.

I would say that TR like many companies focus more on the “backend” stuff than the app. Just think of e. g. the Netflix app and your relation to Netflix, or Amazon’s Prime Video app and your relation to Amazon: they put a lot of effort into the content-side of things, on the backend (e. g. with their recommendation engines), but their apps are not always great (especially Amazon’s app). I believe that this is because e. g. Netflix sees the majority of its value lying with its content and not with the app.

1 Like

What’s crazy is that a basic app that displays blue bars consumes 10-15% of a super high-performance computer’s resources. It’s a missed opportunity. Instead they wasted resources on following the competition (Xert) with AI FTP and pushed some overhyped marketing scheme. Eventually, Electron will fall out of favor and TR will be on the back foot scrambling to provide a proper app :man_shrugging:

If I remember correctly, the last major app revamp and facelift was moving to Electron. That was at least three or four years ago. I thought it was a great app both before and afterwards. As such, I’m confident that TR is planning or already has planned their roadmap and will not be scrambling on the back foot.

I haven’t looked at CPU consumption on my laptop but let’s use your numbers. If it were 5-10%, would you still call it crazy? What’s the threshold here for okay vs crazy?

p.s. nice avatar pic. in another lifetime, I used to do track days at VIR when I lived stateside in DC.

I’m not a developer but I am in the industry and I can say with informed confidence gained from 20 years of experience that one in a zillion people would choose a reduction in cpu cycles from 10-15% to say 5% over a feature they probably will use. The cpu utilization part of your argument is a pet peeve affecting nobody.

6 Likes

It doesn’t matter. The new machines have such headroom I could run a huge photoshop batch operation, zwift in max mode, combine and export files in quicktime player, run TR, and still likely not max out all 8 cores of a new 16. This is something that annoys you but it’s like when I see men wearing a scarf and it’s almost 60 degrees. It’s more my problem than a real problem.

3 Likes

I don’t think this argument is very convincing: if you phrase it in terms of CPU cycles, then yes, I think nobody would care. But if you phrase it in terms of battery life or cost of data (cross-platform apps are usually significantly larger because they need to include all the cross-platform libraries), people start to care. I care that e. g. when I run zoom on my 8-core MacBook Pro, the fans spool up and sound like a hair dryer. And infuriating, because neither Skype nor FaceTime waste my CPU cycles like that. Not great if the main job of the app is listening to speech or speaking yourself.

When I lived in Toronto in 2013–2015, I had a cell phone contract with 1 GB data volume included. I tried to be super frugal and only check the news on the subway, but for some strange reason I could never make it last for a month. Turns out that loading common news sites, just the home page, weighed in at up to 30 MB. Load times were not great, too, for obvious reasons. Even at home, I’d have a data cap of initially only 180 GB or something. (WTF Canada, everywhere else in the world you don’t meter data volume on home internet connections.) I also accidentally updated apps once, and FaceBook’s app update ate up one quarter of my monthly data allotment.

So I think it all depends on how you phrase the question. Most people do care about some of these effects.

I’m not so sure. The update took away many basic features I used regularly, including e. g. being able to check whether I hit my power numbers and add annotations right from the app. These are rather basic features in my book (and independent of whether TR relies on native APIs or a cross-platform framework like Electron).

I hear where you’re coming from. But I’m still not sure I buy this argument. Like what Oreo said, when a key feature like group workouts with video maxes out the CPU and dims the lights in the house, that’s a problem. If I were a TR developer working on group video workouts and had to decide between Electron which runs the CPU at 100% and sometimes crashes the app/machine and native runs the CPU at 50% and no crashes, I’d gladly make that switch to native. A poor user experience negates that cool new feature.

That’s exactly what happened with group workouts in my experience. I had to run my machine (an Intel 13” MBP) on battery because I was in the garage. Friends and family were also in similar situations. I couldn’t make it through an hr workout, the app crashed frequently, and was frustrating to use because of this. Our group training fizzled out. It’s great in theory, but in practice was subpar.

Maybe if they ran a native app, I wouldn’t have run into these issues and we’d still be using it. I mean does anyone still do group workouts now? How much money and resources did they throw at that? It’s been 2 years since group workouts were announced and you still can’t do them on iOS or iPadOS :man_shrugging:

1 Like