Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Micron Announces Exit from Crucial Consumer Business

investors.micron.com

Post reply on HN
Print view View original

371–380 of 407 posts

Re: Micron Announces Exit from Crucial Consumer Business

#371

Post by AnthonyMouse »

nextaccountic wrote: post #292
AnthonyMouse wrote: post #243

Earlier quoted context omitted.

Apple puts the RAM in the chip package because they integrate the GPU, and then they want to be able to have multiple channels to feed the GPU without having that many slots. (Their entry level models also don't have any more memory bandwidth than normal PC laptops and there is no real reason they couldn't use a pair of SODIMMs.) But low end iGPUs don't need a lot of memory bandwidth (again witness Apple's entry leve…

> integrating high end GPUs makes you thermally limited. Even if you have a surface area equivalent to a high end cpu and high end gpu, combined in a single die?

A high end CPU (e.g. Threadripper) is 350W. A high end GPU (e.g. RTX 5090) is 575W. That's over 900W. You're past the point of die area and now you're trying to get enough airflow in a finite amount of space without needing five pounds of copper or 10000RPM fans.

Separate packages get you more space, separate fans, separate power connectors, etc.

In theory you could do the split in a different way, i.e. do SMP with APUs like the MI300X, and then you have multiple sockets with multiple heatsinks but they're all APUs. But you can see the size of the heatsink on that thing, and it's really a GPU they integrated some CPU cores into rather than the other way around. The power budget is heavily disproportionately the GPU. And it's Enterprise Priced so they get to take the "nobody here cares about copper or decibels" trade offs that aren't available to mortals.

Re: Micron Announces Exit from Crucial Consumer Business

#372

Post by kccqzy »

donatj wrote: post #83

This seems like a very short-sighted mistake... The AI memory bubble is going to bust eventually. They have such enormous brand recognition and are known for selling a quality-first product. To just throw that all away is insane. For the last ten plus years I have only bought Crucial RAM. I am well beyond disapointed to hear this.

Agreed. The RAM in my current PC is 64GB of DDR4 from Crucial. Now I seriously don’t know which brand I can trust when I build my next PC.

Re: Micron Announces Exit from Crucial Consumer Business

#373

Post by snuxoll »

babypuncher wrote: post #360
snuxoll wrote: post #99

Earlier quoted context omitted.

The literal multi-million dollar question that executives have never bothered asking: When is it enough? Do I, as a player, appreciate the extra visual detail in new games? Sure, most of the time. But, if you asked me what I enjoy playing more 80% of the time? I'd pull out a list of 10+ year old titles that I keep coming back to, and more that I would rather play than what's on the market today if they only had an ac…

Well in the case of Doom: The Dark ages, it's not just about about fidelity but about scale and production. To make TDA's levels with the baked GI used in the previous game would have taken their artists considerably more time and resulted in a 2-3x growth in install size, all while providing lighting that is less dynamic. The only benefit would have been the ability to support a handful of GPUs slightly older than t…

I'm not even talking about RT, specifically, but overall production quality. Increased texture detail, higher-poly models, more shader effects, general environmental detail, the list goes on.

These massive production budgets for huge, visually detailed games, are causing publishers to take fewer creative risks, and when products inevitably fail in the market the studios get shuttered. I'd much rather go back to smaller teams, and more reasonable production values from 10+ years ago than keep getting the drivel we have, and that's without even factoring in how expensive current hardware is.

Re: Micron Announces Exit from Crucial Consumer Business

#374

Post by Felger »

tuetuopay wrote: post #326
Felger wrote: post #128

Earlier quoted context omitted.

Agreed, the first gen MX500 with M3CR023 fw proved IMHO to be the second most reliable SATA SSD 2.5" form factor with the Samsung 860 range SSDs (860 Evo / Pro). Sadly, the MX500 is now difficult to find in western europe. Only lower grade BX500, still quite reliable but not as fast as the MX500 with cache + DRAM. Had quite a lot of controller issues (become sluggish for periods of time) with the sandisk/WD ones like…

Huh, I've had the opposite experience with the BX500. Hit a bit hard on them and the SSD drops off the bus. Or fill them up to 80% and witness them crawl to a stop. Dirt cheap drives, but don't ask too much of them.

I wrote the BX500 ils easier to find than the MX500, not that it is better. Obviously the BX is worse than the MX, having no SLC cache.

Re: Micron Announces Exit from Crucial Consumer Business

#375

Post by EasyMark »

sgnelson wrote: post #93

I feel like the "democratization of technology" is on the back slide. For the longest time, we had more and more access to high end technology at very reasonable price points. Now it feels like if you're not Facebook, Google, OpenAI, etc. etc. computation isn't for you. I hope this is just a blip, but I think there is a trend over the past few years.

It's likely that all the mega cloud and AI companies want regular people forced to go to them for solutions and buying up any companies that might pose a potential for allowing that. In response they will use a small percentage of the trillions being thrown at them to eliminate those companies that allow for self hosting or mid tier providers to thrive.

Re: Micron Announces Exit from Crucial Consumer Business

#376

Post by EasyMark »

thewebguyd wrote: post #105
sgnelson wrote: post #93

I feel like the "democratization of technology" is on the back slide. For the longest time, we had more and more access to high end technology at very reasonable price points. Now it feels like if you're not Facebook, Google, OpenAI, etc. etc. computation isn't for you. I hope this is just a blip, but I think there is a trend over the past few years.

I also hope its just a blip, but I don't actually think it is. The democratization of technology was something that had the power to break down class barriers. Anyone could go get cheap, off the shelf hardware, a book, and write useful software & sell it. It became a way to take back the means of production. Computing being accessible and affordable for everyone = working class power. That is why its backsliding. Tho…

I don't see how it can be a blip if AI actually turns out to be successful. They'll likely gobble up any lose hardware for their datacenters until only scraps are left or the AI bubble pops if AGI isn't achieved in the next few years and stock values fall off a cliff

Re: Micron Announces Exit from Crucial Consumer Business

#377

Post by angoragoats »

seanmcdirmid wrote: post #349
angoragoats wrote: post #331

Earlier quoted context omitted.

Hard disagree on it working well for local AI - all the memory bandwidth in the world doesn’t matter when the GPU it’s connected to is middling in performance compared to dedicated options. Give me one (or several) 3090/4090/5090 any day of the week over a Mac.

I’ve got an M3 Max with 64G, and can run larger models well than a single 5090. Yes, the GPU isn’t as fast, but I have a lot more memory and my GPUs still don’t suck that badly.

You illustrated my point exactly: yes, a single 32GB 5090 has half the memory of your Mac. But two of them (or three 3090/4090s) have the same total memory as your Mac, are in the same ballpark in price, and would be several times faster at running the same model as your Mac.

And before you bring up the “efficiency” of the Mac: I’ve done the math, and between the Mac being much slower (thus needing more time to run) and the fact that you can throttle the discrete GPUs to use 200-250W each and only lose a few percent in LLM performance, it’s the same price or cheaper to operate the discrete GPUs for the same workload.

Re: Micron Announces Exit from Crucial Consumer Business

#378

Post by seanmcdirmid »

angoragoats wrote: post #377
seanmcdirmid wrote: post #349

Earlier quoted context omitted.

I’ve got an M3 Max with 64G, and can run larger models well than a single 5090. Yes, the GPU isn’t as fast, but I have a lot more memory and my GPUs still don’t suck that badly.

You illustrated my point exactly: yes, a single 32GB 5090 has half the memory of your Mac. But two of them (or three 3090/4090s) have the same total memory as your Mac, are in the same ballpark in price, and would be several times faster at running the same model as your Mac. And before you bring up the “efficiency” of the Mac: I’ve done the math, and between the Mac being much slower (thus needing more time to run)…

I don't know. Can you bring your GPUs on an inter-continental plane trip and play with LLMs on the plane? It isn't really that slow for 70B 4-q models. These are very good CPU/GPUs, and they are only getting better.

Re: Micron Announces Exit from Crucial Consumer Business

#379

Post by barrkel »

abustamam wrote: post #350
barrkel wrote: post #342

Earlier quoted context omitted.

I feel this vibe. It's part of the reason I invested in a monster home workstation - threadripper 9995wx, 768gb ecc ram, 96gb Blackwell pro. I expect it may easily be the last proper home PC I buy that is scaled in line with the scaling that I grew up with in the 80s and 90s. Increasingly, what we have are mobile terminals - possibly with a dock for monitor connections - for remote big iron. And the continuous push f…

Whoa that definitely sounds like a monster rig! Out of curiosity, how much did that cost? Blackwell alone can be $10k+! It's definitely an investment, especially since it may soon become a relic, not in terms of specs, but in terms what manufacturers will actually end up making.

A little over 20k CHF. I delegated building (and ensuring the parts work together and testing) to Dalco, a Swiss company that specializes in academic compute clusters and the like. The price was competitive with building it myself.

The CPU is more expensive than the workstation blackwell card. 8x 96GB DIMMs - 96GB was at a corner in the price per GB, 128GB was more expensive per GB - is also more expensive now than the GPU. In fact the prices for that kind of package on ebay seem to be approaching the price of the entire box.

Re: Micron Announces Exit from Crucial Consumer Business

#380

Post by angoragoats »

seanmcdirmid wrote: post #378
angoragoats wrote: post #377

Earlier quoted context omitted.

You illustrated my point exactly: yes, a single 32GB 5090 has half the memory of your Mac. But two of them (or three 3090/4090s) have the same total memory as your Mac, are in the same ballpark in price, and would be several times faster at running the same model as your Mac. And before you bring up the “efficiency” of the Mac: I’ve done the math, and between the Mac being much slower (thus needing more time to run)…

I don't know. Can you bring your GPUs on an inter-continental plane trip and play with LLMs on the plane? It isn't really that slow for 70B 4-q models. These are very good CPU/GPUs, and they are only getting better.

Sure, the GPUs sit in my basement and I can connect to them from anywhere in the world.

My point was not that “it isn’t really that slow,” my point is that Macs are slower than dedicated GPUs, while being just as expensive (or more expensive, given the specific scenario) to purchase and operate.

And I did my analysis using the Mac Studio, which is faster than the equivalent MBP at load (and is also not portable). So if you’re using a MacBook, my guess is that your performance/watt numbers are worse than what I was looking at.

Post reply on HN