Not an apples to apples comparison#. iOS uses way less ram than android, plus has memory compression. That’s why Apple gets away with less ram and much smaller batteries with equal or better performance.
FYI, Android has had zRAM support since KitKat, which is from (checking notes) 2013. Same year as iOS.
iOS uses way less ram than android
Common measurements I have seen is around 40%, I wouldn't say way less, but it is definitely less. Still, 3x more for a model in the same year is impressive (and more than needed to be competitive with iOS) and we should give OnePlus credit for it.
Sadly a lot of low-ish to midrange phones are going back or sticking to 6-8GB today, thanks to the RAM squeeze and the efficiency of iOS is certainly helping Apple here. Certainly nobody is going to complain about the performance of the iPhone 17, despite only having 8GB RAM.
In Android, Samsung doesn't seem to suffer as much. You can pick up an S26 here with 12GB RAM and 256GB storage for 623 Euro, which is a nice midrange price. I guess there are benefits if you can produce your own memory.
Apple is already past that, with the entry-level 17e starting at 700 Euro. The midrange regular 17 starts at 950 Euro. And according to rumors those prices are likely to increase in spring if not earlier.
What's funny is that due in part to minimum wage laws (which caused some of the inflation) the lowest quintile is arguably the one who's kept up with inflation better than the rest, meaning there has been an improvement in the wealth inequality department. But nobody talks about it. Meanwhile I've lost like 20% of my buying power to inflation in the past 3-4 years even taking the meager raises into account.
Maybe it's not technically a "swap system" but I'll be shocked if they're not doing effectively the same thing. How could you avoid having active uncompressed pages and inactive compressed pages? The amount of tracking you'd need to do to use compressed pages directly seems like an impossible amount of overhead.
Can you link one? A quick search just showed me information about compressing inactive apps and that requires very bulky actions with a lot of overhead, much more than going per page. Plus information about compressing inactive pages in a way that's basically the same as swap. Is there a third method I'm not aware of?
> iOS uses way less ram than android, plus has memory compression. That’s why Apple gets away with less ram and much smaller batteries with equal or better performance.
Yeah, by aggressively purging all my Safari tabs (and Apps) unexpectedly, during the most convenient times, like when I'm going through a tunnel on the train and my cellular connection is at its best. Still infuriatingly common on my 17 Pro with 12GB of RAM.
iPhones from that era constantly booted apps out of RAM when you toggled between them, making it really slow to have more than a few apps open at the same time.
These issues pretty much never happened with these big RAM android phones.
I know because I had both and it was night and day. Like going from a netbook to a desktop machine.
Careful. If you acknowledge this provable fact you just invite hordes of specs-table-rotted gigahertz-chasers to post replies that completely miss the point.
- it’s functional making it much easier to reason about for LLMs (eg no side effects)
- it’s compiled (including the html template), so the LLM get instant feedback if something is off and can fix it quicker
- it’s more concise than other frameworks (language, framework, no front/backend split) and consequently you hit the max token limit much less frequently
- it has excellent tailwind support
I told a business partner to just use claude for certain tasks a year ago, and it failed miserably in python where it succeeded in elixir/phoenix liveview. This was a regular occurrence.
LLMs have obviously progressed since, but the principle still stands that they work much better in elixir than python.
So in short: it’s great and arguably better than most other languages and frameworks
I’d like some confirmation if possible, but my gut says the general internet (and therefore the training data) probably has a lot of sloppy Python code along with a lot of sloppy writing about Python whereas Phoenix has majority good code and well-reasoned writing
I disagree with that conclusion. I see tailwind as a cleaner more succinct version of css that is much easier to manage and add features too.
Sure it’s not as dry, but I’ve been bitten in this regard because css framework and templates are so intransparent, preventing me from simply changing padding or margin.
CSS is too detailed and too verbose. Frameworks like bootstrap are too high level and don’t give enough control. Tailwind hits the sweet spot whilst allowing me to be detailed if I want to. It allows me to just get it done.
> Sure it’s not as dry, but I’ve been bitten in this regard because css framework and templates are so intransparent, preventing me from simply changing padding or margin.
How does this happen? You can always override css values. Either by ordering, !important, inline or, to make very sure, with inline !important.
I found that once I started needing to use inline styles or !important or in-line important, I was in for a world of hurt and felt sorry for whoever maintained the project after me.
Google's lack of customer service isn't new or limited to GCP. They also don't provide any human help if you're an advertiser with them unless you spend a crazy amount of money. Twenty years ago I used to spend upwards of $20-30k a month with them and I couldn't get a single reply to any inquiry I ever sent.
If you spend $XXX million / year with them on GCP they will, however, assign a person to be your main point of contact.
How does it work if the contract hasn't been fulfilled by Tesla? Wouldn't that nullify the entire contract and Tesla would have to return the full price paid and presumably collect the car.
Munich is a bad example - they were effectively „bought out“ by Microsoft by investing hugely into the local economy in the form of offices and employees. It was also two parties that kept flip flopping with different priorities.
Linux itself had some hiccups but was fine from what I recall.
Yeah, let me dispute that. They were, at least on three occasions, forced to roll back due to "citizen sent me X and can't open it" and/or "sent Y to citizen and they can't open it" concerns.
Mind you: these issues still persist in a fully Microsoft/Adobe "solution environment", but less so than in the "disregard all and move to Linux" situation.
And to be perfectly clear: that's all unacceptable. But it adds another, say, EUR 2B to the equation.
Had similar thoughts until googling what the cheapest Chromebook on Amazon costs: $139.
I’d still get a Neo and for students it’s probably the right choice - chromebooks are just a browsers after all.
But pricing wise this laptop is a decade too late. The netbook of the day (chromebooks?) are just unbelievably cheap.
Apple will still sell millions of these and keep on eating up market share.
Tidewave.ai does exactly that. It’s made Claude code so much more functional. It provides mcp servers to
- search all your code efficiently
- search all documentation for libraries
- access your database and get real data samples (not just abstract data types)
- allows you to select design components from your figma project and implements them for you
- allows Claude to see what is rendered in the browser
It’s basically the ide for your LLM client. It really closes the loop and has made Claude and myself so much more productive.
Highly recommended and cheap at $10/month
Ps: my personal opinion. I have Zero affiliation with them
They do bin their chips. Across the range (A- and M-series) they have the same chip with fewer / disabled cpu and gpu cores. You pray a premium for ones with more cores. Unsure about the chip frequencies - Apple doesn’t disclose those openly from what I know.
I’ve been using this since early this year and it’s been great. It was what convinced me to just stick to Postgres rather than using a dedicated vector db.
Only working with 100m or so vectors, but for that it does the job.
The biggest selling point to using Postgres over qdrant or whatever is that you can put all the data in the same db and use joins and ctes, foreign keys and other constraints, lower latency, get rid of effectively n+1 cases, and ensure data integrity.
I generally agree that one database instance is ideal, but there are other reasons why Postgres everywhere is advantageous, even across multiple instances:
- Expertise: it's just SQL for the most part
- Ecosystem: same ORM, same connection pooler
- Portability: all major clouds have managed Postgres
I'd gladly take multiple Postgres instances even if I lose cross-database joins.
Yep. If performance becomes a concern, but we still want to exploit joins etc, it's easy to set up replicas and "shard" read only use cases across replicas.
Depends on the query and I don’t have exact numbers of the top of my head, but we’re talking low 100ms range for something pgvector itself wasn’t able to handle in a reasonable amount of time.
# pun intended
reply