dampfnudel
macrumors 603
I hope we get back to normal by this time next year. I would like to buy a SSD at a normal price and I don’t want Sony to delay the release of the PS6.
That's now how contracts work. The contract on the supply run might be up for the current models. Time for the new ones.This is the demonstration that money cannot buy everything. When there is no memory chip available, there is none. At least for a time. There are things that are on limited supply, be careful with these.
You'd want more memory to run good, local models - at least if you want to run other applications as well, e.g. the ones using the AI. I'd get at least 48 GB to run the better versions of Gemma4, which is a good, efficient, and recent one.Ironically, I am thinking about buying a basic 24 GB M5 Pro to experiment with local AI models...
What would you recommend? I am a researcher but not in a computing heavy area (economics). I do wanna use local models to help me with the job though. 48 GB?You'd want more memory to run good, local models - at least if you want to run other applications as well, e.g. the ones using the AI. I'd get at least 48 GB to run the better versions of Gemma4, which is a good, efficient, and recent one.
At least 48 GB, but preferably 64 GB. Of course, more is better but Apple charged an arm and a leg for memory even before prices increased. It also depends on the configurations available, sometime one CPU configuration goes 24 - 48 - 96 GB while another goes 32 - 64 - 128What would you recommend? I am a researcher but not in a computing heavy area (economics). I do wanna use local models to help me with the job though. 48 GB?
The M4 Pro mini goes 24 48 64, 12 core cpu and 16 core gpu or 14 cpu 20 gpu. Suppose the M5 will be the sameAt least 48 GB, but preferably 64 GB. Of course, more is better but Apple charged an arm and a leg for memory even before prices increased. It also depends on the configurations available, sometime one CPU configuration goes 24 - 48 - 96 GB while another goes 32 - 64 - 128
Good. I am in the market for a Mac mini. Will not buy the m4 pro but wait to pounce as soon as pre orders for the m5 openI can order a 16” MBP M5 Max with 128GB RAM with delivery for April 24th so it’s definitely not memory related and more due the incoming M5 Mac Studio and Mac mini refresh.
This makes absolutely no sense as the M6 will likely be debuting around that same time. More likely is the Mini might skip the M5 entirely if it's not updated soon similar to how it skipped the M3.Mac mini with M5 and M5 Pro chips at some point in September or October this year.
That's not an either/or question.Is AI an ‘overhyped’ fad as many people seem to think? No. It’s a major tech revolution that, for good or for bad, is going to change life on earth forever.
It has nothing to do with this.This is the demonstration that money cannot buy everything. When there is no memory chip available, there is none. At least for a time. There are things that are on limited supply, be careful with these.
At least 48 GB, but preferably 64 GB. Of course, more is better but Apple charged an arm and a leg for memory even before prices increased. It also depends on the configurations available, sometime one CPU configuration goes 24 - 48 - 96 GB while another goes 32 - 64 - 128
Then this is a horrible way to handle it. You stop production for a supposed WWDC release in June? People who need a system now are going elsewhere. Apple continues to handle their pro markets horribly. Between how they botched the Mac Pro since 2013, and now this. At this point I am just going back to Windows.Macrumors has it wrong. PCs remain available. This has to be the M5 refresh.
People also don't understand if the bubble bursts, the big push from several CEOs are that they want us to start renting cloud hardware and we will own nothing.We see variations on this "I wish this brainless bubble would hurry up and burst. Nobody is too big to FAIL" kind of comment with some frequency. I would love to have someone explain the thinking to me.
Because it seems to me that AI is a solid evolutionary tech concept still in its early years. No doubt some of the paths that AI growth may take will prove unsuccessful, "fail," but that is just the way growth works. IMO AI is not going to fail even if some current growth directions prove to have been bad choices financially for the folks wealthy enough to be investing in option A versus option B versus option C...
I decided it was the perfect time to sell my M4, Mac mini Pro and I got a pretty good price for it. Sold in less than one day. I just wasn't using it. I still have a MacBook Air and a MacBook Pro and I use those more than anything.
I hope we get back to normal by this time next year. I would like to buy a SSD at a normal price and I don’t want Sony to delay the release of the PS6.
The bubble, over-hyping and unsustainable levels of investment has nothing directly to do with the technology itself going away. If the bubble bursts, either a few big players will remain standing or new, more realistic companies making more realistic claims will pick up the tech.
Are some early investors in AI going to fail? It certainly looks that way. But some will succeed, and as always, a few filthy rich billionaires will get even filthier from it.
There's more to a "bubble" than some early investors failing. That happens with virtually every development and it's often the second mouse that gets the cheese. When something like the AI bubble happens it distorts the markets and disrupts other businesses. We're seeing this with the current RAM and SSD shortage/price rises which are already causing damage. It's really not going to help the AI industry if people can't afford the phones and computers needed to access their services! The gaming industry is being harmed by the lack/high price of GPUs and VRAM. Huge software development efforts are going in to AI features that customers are rejecting - some are starting to backpedal (https://www.xda-developers.com/microsofts-windows-11-copilot-purge-has-already-started/). Existing non-AI companies/products are being bought out for ridiculous prices because someone thinks they can be pivoted into subscription-based AI cash registers (see: Affinity Studio). AI products are being pushed out prematurely, failing and discrediting the tech (https://www.bbc.co.uk/news/articles/cq5ggew08eyo).You actually seem to be agreeing with what I said in the part that you didn’t quote:
...
Are some early investors in AI going to fail? It certainly looks that way. But some will succeed,
You are trying to dismiss the existence of the AI bubble and the chaos it is already causing as just another case of there always being winners and losers in business. There is a lot more to the current AI bubble than that.Am I misunderstanding your argument?
Heya, is there a ‚solid‘ middle-ground for something like roleplay or fanfiction/story chatbots? I agree that 24GB are pretty low, but going 128GB is costly. Are there solid models that would harmonize well with 48 or 64GB? I own a 32GB M1 Max and consider upgrading to a M5 Pro or Max soon. Thanks!That's what I did. Let me tell you--the local AI models when you use 24 gb of RAM are terrible. They get most answers wrong. It will look like you are running ChatGPT or whatever locally, but its knowledge is trash. Wasn't worth it for me, and I'm told you need at least 100 gb to get an experience approaching the paid, cloud models.
I have the 24gb M4 and while you can load some decent models with 24gb you basically have no memory left to actually use them all that much.Ironically, I am thinking about buying a basic 24 GB M5 Pro to experiment with local AI models...