Earlier quoted context omitted.
Cache is not general purpose as far as I know. I want to be able to do whatever I want with it.
Where’s the market demand for that?
Where is the market demand for faster RAM? This isn't a good question.
How DRAM changed the world
91–100 of 128 posts
Re: How DRAM changed the world
#92The article doesn't properly explain how DRAM is different from SRAM. DRAM has to constantly refresh itself in order not to 'forget' its contents.
Why would we use DRAM, then? It seems better not to have to refresh it all the time. (I think I more or less know, but I’d rather talk about it than look it up this morning.)
The key point is that the refreshes do not need to happen very often. Something like once per 20 ms for each row was doable even by an explicit loop that the CPU had to periodically execute.
And this task soon moved to memory controllers, or at least got done by CPUs automatically without need for explicit coding.
Re: How DRAM changed the world
#93Earlier quoted context omitted.
5nm can hold roughly a gigabyte of SRAM on a cpu-sized die, that's around $130/GB I believe. At some point 5nm will be cheap enough that we can start considering replacing DRAM with SRAM directly on the chip (aka L4 cache). I wonder how big of a latency and bandwidth bonus that'd be. You could even go for a larger node size without losing much capacity for half the price.
SRAM also requires more power than DRAM and the simple regular structure of SRAM arrays compared to (other) logic makes it possible to get good yield rates through redundancy and error correction codes so you could have giant monolithic dies, but information can't exceed the speed of light in a medium. There just isn't enough time for the signals to propagate to get the latency you expect of a L3 cache out of gigabyt…
Oh yeah this would definitely be something like L4 cache rather than L3 like AMD's X3D cpus. The expectation is as an alternative to DRAM (or as a supplement), kind of like what Xeon Phi did.
Re: How DRAM changed the world
#94Earlier quoted context omitted.
5nm can hold roughly a gigabyte of SRAM on a cpu-sized die, that's around $130/GB I believe. At some point 5nm will be cheap enough that we can start considering replacing DRAM with SRAM directly on the chip (aka L4 cache). I wonder how big of a latency and bandwidth bonus that'd be. You could even go for a larger node size without losing much capacity for half the price.
5nm will never be that cheap. The performance benefit would be easily 2x or more though.
Even 15 years from now?
Re: How DRAM changed the world
#95The article doesn't properly explain how DRAM is different from SRAM. DRAM has to constantly refresh itself in order not to 'forget' its contents.
It isn't the point of the article, but this is true of every storage medium. It's just a question of milliseconds or years.
Static RAM ( based on how it is used) never needs to be refreshed at current typical computer power on times (hours or days ) . Current DRAM must be refreshed at very much faster rates to be able to be useful.
Re: How DRAM changed the world
#96Earlier quoted context omitted.
Years ago. DRAM uses a capacitor. Those capacitors essentially hit a hard limit at around 400MHz for our traditional materials a very long time ago. This means that if you need to sequentially read random locations from RAM, you can't do it faster than 400MHz. Our only answer here is better AI prefetchers and less-random memory patterns in our software (the penalty for not prefetching is so great that theoretically l…
What does “faster than 400MHz” mean in this context? Does that mean you can’t ask for a unit of memory from it more than 400M times a second? If so, what’s the basic unit there, a bit? A word? I built a little CPU in undergrad but never got around to building RAM and admit it’s still kind of a black box to me. Bonus question: When I had an Amiga, we’d buy 50 or 60ns RAM. Any idea what that number meant, or what today…
The capacitors take time to charge and discharge. You can't do that more than around 400MHz with current materials. You are correct that it means you can't access the same bit of memory more than 400M/sec. This is the same whether you are accessing 1bit or 1M bits because the individual capacitors that make up those bits can't be charged/discharged any faster.
When we moved from SDR to DDR1, latencies dropped from 20-25ns to about 15ns too, but if you run the math, we've been at 13-17ns of latency ever since.
Re: How DRAM changed the world
#97I have a recollection of a design where microprocessor reads were used to refresh DRAM contents. Late 1970s. I thought it was in a early 6800 Motorola book. Can find it now, or no mention of the technique, now. Would slow down program operation for sure. Maybe my recollection is wrong, not sure.
Re: How DRAM changed the world
#98Earlier quoted context omitted.
The problem with 4k, which 99% of people only experience as a stream, is that it's over-compressed. A 1080p BluRay is 36 Mbit. Netflix 4k is 15 Mbit. So unless I see people mentioning the media, I am always weary of the comparison.
Why stream when you can play files from a local media server? Cheaper and I'm sure the Mbit rate is better.
Because the majority of people don't do that.
Re: How DRAM changed the world
#99Earlier quoted context omitted.
If that's the case, why haven't we switched to SRAM? Isn't it only about 4x the price at any given process node?
That 4x the price does also explain why it has not happened.
If it were even 20% faster than DRAM, there would be a market for it at the higher price. The post I replied to was asserting that there was a physical limit of 400MHz for DRAM entirely due to the capacitor. If SRAM could run with lower latency, memory-bound workloads would get comparably faster.
Re: How DRAM changed the world
#100Earlier quoted context omitted.
It's insufficient for large theaters and barely sufficient for medium theaters. 1080p is plenty for most home theaters (though with the so-so quality of streaming, I wonder to what degree macroblocks are the new pixels)
1080p looks different to 4k. Just because its 'plenty' doesn't mean 4k doesn't matter. And we are at 2024, i own a 4k lg oled display now for years. Why not leverage it? Just because 1080p is 'plenty'?
With approximately 90% of the HT setups[1] I have seen, I can downscale a 4k video to 1080p and have the owner of the HT setup not be able to successfully ABX the difference.
1: I'm using the term "HT setup" rather broadly, as the primary location in a residence for watching movies as a group; it includes e.g. people who don't own a TV and watch movies on their laptop sitting on a coffee table. Setups where the display covers over 40% of the FOV (where 4k definitely makes a difference) are somewhere in the top 5%.