474 karma · joined September 1, 2020
There was a simple function in java standard library which was wrong for years because of this phenomenon.
https://dev.to/matheusgomes062/a-bug-was-found-in-java-after...
That's like trying to plug a fist-sized leak with a toothpick.
It's not nothing, but it's not a good excuse to waste more single-use plastic (as you're implying here saying "Single-use plastic waste is 'good'")
These people are awful.
That's like saying you should never use anything other than RAID0 because anything else costs you more for the same performance. Yeah, sometimes cost/performance aren't the only important metrics!
The pandemic has literally been a real-life example of the downsides of the approach he is advocating.
Yes, "CUDA Core" is (was?) analogous to "FP32 ALU" - (it may have transitioned to FP16 nowadays to double the count?)
Yes, EPYC has 16 FP32 ALU per Core.
But, single GPU "Compute Units" often have 2-4x that (up-to 64 FP32 ALU), and a lot of other differences which when added up at the whole GPU level don't mean they max out anywhere near '64 "cores"' "if you counted a GPUs independent parallel threads of execution like on CPUs"
That's where they couldn't be more different!
Also, for accuracy, vis-a-vis "independent parallel threads of execution like on CPUs", even the 64-core Threadripper has 128, not 64.
And many GPUs with 64 physical "Compute Units" would have at least 640 by the same measure (since many have 10 independent programs which can be resident in hardware at once, compared to SMT2 on Threadripper).
But when you break it down further, while a model GPU Compute Unit from a few years ago might have 64x FP32 ALUs in hardware, that CU has the capacity to schedule and execute single instructions on 10*64 logical threads (= 640 logical threaded instructions in-flight on a single CU at once). On a 64CU chip, that's 40,960 logical floating point instructions in flight at once, each one of these coming from a different logical thread of execution in a SIMT model.
The superscalar CPU cores can also have lots of instructions in flight, but they are more deeply pipelining instructions for fewer threads of execution (focused on single-threaded performance).
This is all a long way of saying that a.) you may not have been so far off when comparing raw ALU counts; but, b.) you couldn't be misrepresenting the facts more when comparing the differences in "parallel threads of execution."
This is a core architectural difference between CPUs and GPUs, and while yes, there are similarities in ALU count, the way the transistors and ALUs are utilized by software is quite different.
To oversimplify, and ignoring performance of applying these different programming models to different compute architectures:
Using an example 64-CU GPU compared to 64-Core EPYC:
64 CU GPU as SIMD: 640 Threads (SMT10) and 4096 FP32 ALUs (SIMD64 - actually 4xSIMD16)
64 Core CPU as SIMD: 128 Threads (SMT2) and 1024 FP32 ALUs (SIMD16 - actually 2xSIMD8)
64 CU GPU as SIMT: 40,960 Threads (10:1 Threads:ALUs) 64 Core CPU as SIMT: 2,048 Threads (2:1 Threads:ALUs)
So, this model 64-CU GPU can schedule 20x more logical hardware threads in a SIMT programming model than the 64-Core EPYC CPU, despite having only 4x the FP32 ALUs.
My final caveat would be that GPUs are over time increasing single-threaded performance, and CPUs are becoming wider. So, in a way, there is some architectural convergence - but it's not to be overstated.
One last note: Niagara was a bit GPU-like, with round-robin thread scheduling, and POWER now has SMT8. But differences remain.
I'm definitely where I am now because of the internet and Wikipedia.
At the same time, I agree with you about how the internet and technology have changed and continue to change. I agree that too much knowledge is still vaulted away, that a few companies have established large monopoly positions in tech, that they are destroying the egalitarian dream of the early internet.
I do recognize, however, that I'm the exception and not the rule in this regard.
The internet now is very different than the internet I grew up on, and it isn't the future I hoped for (where I imagined everyone would learn as I did, directly benefiting from having access to much of humanity's knowledge).
I lament what the internet has turned into - the overwhelming success of centralized platforms, surveillance capitalism, viral misinformation, etc. I worry about how much worse this all may get once deep-fakes become commonplace...
P.S. I also played video games - but overall I spent more time reading and learning/building things than playing games.
It's meaning in the cultural zeitgeist, however, has evolved to include people who don't deny that the climate does change over time, or even is currently changing; but, who deny the science supporting anthropogenic climate change, specifically the clear impacts to atmospheric CO2 starting during the industrial revolution and driven primarily by fossil fuel emissions.
Essentially, I think most people would agree at this point that the common usage also includes people who deny we can/should do anything about climate change, or that the current situation vis-a-vis greatly accelerated climate change is our "fault."
The records re: CO2 are not hard to find, they come from ice cores and date back ~800,000 years.
You can start here, but there are plenty of independent resources online: https://www.climate.gov/news-features/understanding-climate/...
What exactly do you think the definition of a "climate change denier" is?
The important part isn't recognizing that climate change exists. The important part is recognizing why and what needs to be done about it.
You're misinformed about the science on this one...