Literally a previous gen card.

the fact that google pay firefox anually meaning that its in best interest of google that there is no monopoly, judge says

[0] https://thisdayintechhistory.com/12/06/apple-sues-over-quick...

Right now if the US wants to go to war with China, or anyone China really really likes, they can expect with high probability to very quickly encounter major problems getting the best chips. AIUI the world has other fab capacity that isn't in Taiwan, and some of it is even in the US, but they're all on much older processes. Some things it's not a problem that maybe you end up with an older 500MHz processor, but some things it's just a non-starter, like high-end AI.

Sibling commenters discussing profits are on the wrong track. Intel's 2024 revenue, not profits, was $53.1 billion. The Federal Government in 2024 spent $6,800 billion. No entity doing $1.8 trillion in 2024 in deficit spending gives a rat's ass about "profits". The US Federal government just spends what it wants to spend, it doesn't have any need to generate any sort of "profits" first. Thinking the Federal government cares about profits is being nowhere near cynical enough.

The AI hype train was built on the premise that AI will progress linearly and eventually end up replacing a lot of well paid white collar work, but it failed to deliver on that promise by now, and progress has flatlined or sometimes even gone backwards (see GPT-5 vs 4o).

FAANG companies can only absorb these losses for so long before shareholders pull out.

To that point, they've been "just getting started" in practically every chip market other than x86/x64 CPUs for over 20 years now, and have failed miserably every time.

If you think Nvidia is doing this because they're afraid of losing market share, you're way off base.

  224 GB/s

  128 bit 

I love GPU differentiation, but this is one of those areas where Nvidia is justified shipping less VRAM. With less VRAM, you can use fewer memory controllers to push higher speeds on the same memory!

For instance, both the B50 and the RTX 2060 use GDDR6 memory. But the 2060 has a 192-bit memory bus, and enjoys ~336 GB/s bandwidth because of it.

My RTX 5090 is about 10x faster (measured by FP32 TFLOPS) and I still don't find it to be fast enough. I can't imagine using something so slow for AI/ML. Only 2.2 tokens/sec on an 8B parameter Llama model? That's slower than someone typing.

I get that it's a budget card, but budget cards are supposed to at least win on a pure price/performance ratio, even with a lower baseline performance. The 5090 is 10x faster but only 6-8x the price, depending on where in the $2-3,000 price range you can find one at.

Yes.

> Other than the market segmentation over RAM amounts, I don't see very much difference.

The difference between CDNA and RDNA is pretty much how fast it can crunch FP64 and SR-IOV. Prior to RDNA, AMD GPUs were jacks of all trades with compute bias. Which made them bad for gaming unless the game is specifically written around async compute. Vega64 has more FP64 compute than the 4080 for context.

I think if AMD was able to get a solid market share of datacenter GPUs, they wouldn't have unified. This feels like CDNA team couldn't justify its existence.

This feels like a 'brand new sentence' to me because I've never met an ALi chipset that I liked. Every one I ever used had some shitty quirk that made VIA or SiS somehow more palatable [0] [1].

> Intel started releasing platforms with very little PCIe connectivity,

This is also a semi-weird statement to me, in that it was nothing new; Intel already had an established history of chipsets like the i810, 845GV, 865GV, etc which all lacked AGP. [2]

[0] - Aladdin V with it's AGP Instabilities, MAGiK 1 with it's poor handling of more than 2 or 3 'rows' of DDR (i.e. two double-sided sticks of DDR turned it into a shitshow no matter what you did to timings. 3 usually was 'ok-ish' and 2 was stable.)

[1] - SIS 730 and 735 were great chipsets for the money and TBH the closest to the AMD760 for stability.

[2] - If I had a dollar for every time I got to break the news to someone that there was no real way to put a Geforce or 'Radon' [3] in their eMachine, I could have had a then-decent down payment for a car.

[3] - Although, in an odd sort of foreshadowing, most people who called it a 'Radon', would specifically call it an AMD Radon... and now here we are. Oddly prescient.

I realize the AGX is more of a low power solution and it's possible that nvidia is still technically limited when building SOCs but this is just speculation.

Does anybody know actual ground truth reasoning why Nvidia is buying Intel despite the fact that nvidia can make their own SOCs?

It also happened under G. W. Bush with banks and auto manufacturers, but the worst offense was under Nixon with his nationalization of passenger rail.

At least with the bank and car manufacturer bailouts the government eventually sold off their stocks, and with the Intel investment the government has non-voting shares, but the government completely controls the National Railroad Passenger Corporation, (the NRPC aka Amtrak) with the board members being appointed by the president of the United States.

We lost 20 independent railroads overnight, and created a conglomerate that can barely function.

If you fiddle and concentrate only on the top performers, the bottom falls out. Most of the US economy is still in small companies.

Let's assume Trump admin pressured Nvidia to invest in intel.

Chips act (voted by Democrats / Biden) gave Intel up to $7.8 billion of YOUR money (taxes) in form of direct grants.

Was it more of "Mussolini-style corporatism" to you or not?

It leads to mistakes like you mention, where a new market segment or new entrant is not a sure thing. And then it leads to mistakes like Larrabee and Optane where they talk themselves into overconfidence (“obviously this is a great product, we wouldn’t be doing it if it wasn’t guaranteed to make $1B in the first year”).

It is very hard to grow a business with zero risk appetite. You can’t take risky high return bets, and you can’t acknowledge the real risk in “safe” bets.

The problem is, console manufacturers know precisely how much of their product they anticipate to sell, and it's usually a lot. The PlayStation 5 is 80 million units so far.

And at that scale, the console manufacturers want to squeeze every vendor as hard as they can... and Intel didn't see the need to engage in a bidding war with AMD that would have given them a sizable revenue but very little profit margin compared to selling Xeon CPUs to hyperscalers where Intel has much more leverage to command higher prices and thus higher margins.

> they turned down Apple on the iPhone

Intel just was (and frankly, still is) unable to compete on the power envelope with ARM, that's why you never saw x86 take off on Android as well despite quite a few attempts at it.

Apple only chose to go for Intel with its MacBook line as PowerPC was practically dead and offered no way to extract more performance, and they dropped Intel as soon as their own CPUs were competitive. To get Intel CPUs to the same level of power efficiency that M-series CPUs have would require a full rework of the entire CPU infrastructure and external stack, that would require money that even Intel at its best frankly did not have. And getting x86 to be power effective enough for a phone? Just forget it.

> Clearly the threat of an Arm or RISC-V finding itself fused to a GPU running AI inference workloads has woken someone up, at last.

Actually, that is surprising for me as well. NVIDIA's Tegra should easily be powerful enough to run the OS for training or inference workload. If I were to guess, NVIDIA wants to avoid getting caught too hard on the "selling AI shovels" train.

They just need to separate business units.

But with the state of the courts today... who knows..

Intel's foundry costs are probably competitive with nvidia too - nvidia has too much opportunity cost if nothing else.

While it doesn't quite compete at performance and power consumption, it does at price/performance and overall value. It is a $250 card, compared to the $300 of the 4060 at launch. You can still get it at that price, if there's stock, while the 4060 hovers around $400 now. It's also a 12GB card vs the 8GB of the 4060.

So, sure, this is not competitive at the high-end segment, but it's remarkable what they've accomplished in just a few years, compared to the decades that AMD and NVIDIA have on them. It's definitely not far fetched to assume that the gap would only continue to close.

Besides, Intel is not only competing at GPUs, but APUs, and CPUs. Their APU products are more performant and efficient than AMD's (e.g. 140V vs 890M).

How was Intel "circling the drain"?

They have a very competitive offering of CPUs, APUs, and GPUs, and the upcoming Panther Lake and Nova Lake architectures are very promising. Their products compete with AMD, NVIDIA, and ARM SoCs from the likes of Apple.

Intel may have been in a rut years ago, but they've recovered incredibly well.

This is why I'm puzzled by this decision, and as a consumer, I would rather use a fully Intel system than some bastardized version that also involves NVIDIA. We've seen how well that works with Optimus.

AMD has always followed closely NVIDIA in crippling their cheap GPUs for any other applications.

After many years of continuously decreasing performance of the "consumer" GPUs, only Intel has offered in the Battlemage GPUs FP64 performance comparable with what could be easily obtained 10 years ago, but no longer today.

Therefore, if the Intel GPUs disappear, then the choices in GPUs will certainly become much more restricted than today. AMD has almost never attempted to compete with NVIDIA in features, but whenever NVIDIA dropped some feature, so did AMD.

Would be foolish to throw that away now that they're finally getting closer to "a product someone may want to buy" with things like B50 and B60.

One wonders just how bad things must have been internally for that to be the state of one of their core IPs in this day and age...

This needlessly divisive and devoid of any factual basis. No gulags will exist and you know it.

There have been a lot of mergers where that has not happened.

This all ignores the near complete lack of product out of their advanced processes as well.