Telegram Desktop Talk

27 June 2025 21:34

Hi I' wondering if i'ts posible to communicate with a signal use through

Telegram Desktop Talk

27 June 2025 19:15

😂 When I click on links of the 🤒 form /channel/keralajs/20463/26116 TDesktop 5.15.4 crashes ☠️ #bug 👽 😤 Crashed with this link 👍 🎉

❌ Clicking on 😮‍💨 message link does not crash if I use 5.12.3 👺

🫡 Does not crash on links of the 🤡 form /channel/TheForum/289/1290 😮

Telegram Desktop Talk

27 June 2025 15:33

This is a very plausible reason 🤔

Telegram Desktop Talk

27 June 2025 08:48

what do I know what's causing that error message. it's not like telegram doesn't obfuscate the real reasons for server-related behavior elsewhere!

Telegram Desktop Talk

27 June 2025 08:47

what do you mean? doesn't error say that it's "currently unavailable"?

Telegram Desktop Talk

27 June 2025 08:28

or they are moving to grok translation or something

Telegram Desktop Talk

27 June 2025 08:28

lmao translations still don't work

Telegram Desktop Talk

27 June 2025 01:20

I guess something irreplaceable must have caught fire or exploded with the translation service

Telegram Desktop Talk

26 June 2025 21:11

https://x.com/telegram/status/1938311068679606595

Telegram Desktop Talk

26 June 2025 21:06

I mean sure, but doesn't that just suggest an if..else block for it on the server?

Telegram Desktop Talk

26 June 2025 21:05

Undefined behavior is bad and exceptions will trickle further down, so it's good that it was considered at the time

Telegram Desktop Talk

26 June 2025 20:56

If there's an error message it means they probably care and are aware of it?

Telegram Desktop Talk

26 June 2025 20:55

Thanks for the information.

Telegram Desktop Talk

26 June 2025 16:10

Here too. TG Direct.

Telegram Desktop Talk

25 June 2025 14:13

hello, desktop chat! what are we upto? anything new and interesting?

Telegram Desktop Talk

27 June 2025 19:53

Used to crash, doesn't crash anymore after fetching the message

Telegram Desktop Talk

27 June 2025 15:47

Now that you mention it… Could be yeah, considering those previous mentions by Durov that he and Musk would pursue further collaboration. Hopefully it’s not a precedent to “move fast and break things”, but I suppose it’s not an easy change either…

Telegram Desktop Talk

27 June 2025 08:49

it still tells you that it's temporarily unavailable

Telegram Desktop Talk

27 June 2025 08:47

which basically means it's temporarily offline

Telegram Desktop Talk

27 June 2025 08:46

I don't need to know the reason but what bugs me the most is that if it wasn't for other users I wouldn't even know that the problem is caused by something server-side

Telegram Desktop Talk

27 June 2025 08:28

maybe they ran out of credits in their google console

Telegram Desktop Talk

27 June 2025 01:33

i am sure telegram have backups of everything, and multiple copies

Telegram Desktop Talk

26 June 2025 21:20

At least they're aware of and working on it, that's good... Hopefully it's nothing too bad. But something like this is what I'm imagining. Reject incorrect input (inner else), and fail gracefully / perform read-only behavior for everything else. Project is still proprietary, so screenshot only and all rights reserved.

Telegram Desktop Talk

26 June 2025 21:07

It's that else, do this in every other condition, that I imagine this to be in for when a connection to a translation service fails or such

Telegram Desktop Talk

26 June 2025 21:05

All apps are showing the same, and it's not related to the clients

Telegram Desktop Talk

26 June 2025 21:02

Eh, I'd imagine that to be more of a "people like Preston cared about it when it was implemented" kind of thing

Telegram Desktop Talk

26 June 2025 20:55

it's like they really don't care

Telegram Desktop Talk

26 June 2025 20:55

no message from telegram about this either

Telegram Desktop Talk

26 June 2025 16:10

are translations disabled for anyone else or just me?

Telegram Desktop Talk

24 June 2025 21:44

There is another thing I would like to talk about, and that is "unused RAM is wasted RAM". Generally it gets me quite violent, I'll refrain from that here. But as someone else replied, that "unused" RAM goes to filesystem cache. And that's orders of magnitude faster (both in bandwidth and latency) than even a premium NVMe drive.

Every PC has CPU cache, RAM, and storage. Those are tiered, in the hopes that the data you want is always as close to the CPU as possible. So that unused RAM that holds cache, is actively making your system more responsive — by a long shot. Even RAM runs circles around any storage media available today. You can make exceptions to that, e.g. contemporary SSDs have reached the bandwidth of DDR(1) RAM (~300MB/s). Maybe NVMe has reached DDR2 by now, not sure but it seems reasonable. But I hope it conveys the orders of magnitude difference that these media have.

There is another way that this argument became prevalent in my watch project. It seems easiest to pick either 8MHz with internal RC (default), or 1MHz for the ATtiny85 chip. At 1MHz the chip would spend about 80-96% time active. Not only is it too close to 100% (limiting expansion and even tuning), but it also negates the power savings that 1MHz would normally be chosen for. 4MHz seemed like a good trade-off, even if that means that 75% of that time would be "wasted". As with everything in engineering, it's a trade-off. And those tend to be a lot more complex than just "not using resource X for purpose Y means it's useless".