HANLON’S RAZOR: Never attribute to malice that which is adequately explained by stupidity.
I try to stick to this where I can. In the case of Nvidia, I can’t see how this can be explained by stupidity, so I propose their intent is malice.
Nvidia made some recent announcements. The RTX 4000 series is coming with RTX 4090, RTX 4080, and RTX 4080. That’s not a typo. There’s 2 different RTX 4080 parts. Additionally, the RTX 6000 was announced. This is also not a typo.
I’ll explain what these are.
The RTX 4090 is the top end gaming graphics card. There are two tiers of RTX 4080, one which will cost around $1000, and another which will cost around $800. The lower cost one will have less video RAM and fewer CUDA cores, so expect lower performance. As for the RTX 6000, this is a workstation part that will cost far more than the RTX 4090, but will likely have lower gaming performance. That one is not intended for the home user. If you don’t know whether you need a workstation graphics card, you probably don’t need it, so save the money.
It’s speculated the lower end RTX 4080 was intended to be RTX 4070 at $800, but with the price crash of graphics cards coming out of the crypto crash, there was possibly a panic over the idea of charging nearly twice the price of a RTX 3070, and definitely around twice the price of a RTX 2070. And the sad thing is, if this is true, that means people who typically go for the **70 spec are going to be given far lesser as an option, and likely for that **70 price tag, if not higher (wouldn’t surprise me to see it around $600 – remember when this tier was like $200-300?).
This is just the start for this new generation. We haven’t seen the lower end parts, yet. We haven’t seen the lower power parts for laptops, yet. That last one is probably going to be a big one. Because in the 3000 series, they did some more fuckery.
There exists RTX 3080, RTX 3080, and RTX 3080 with Max-Q. Seems familiar, right? I just pointed out there’s 2 RTX 4080 parts, so this problem started in the 3000 series, then, right? Well, I don’t know when this started, but there’s a different problem with this one. One of those non-Max-Q RTX 3080 is using a marketing name, while documentation will refer to it as a Laptop part. But it bears the same name. This persists with the RTX 3080 Ti, RTX 3080 Ti, and RTX 3080 Ti with Max-Q, and even LTT got in on perpetuating the confusion . Yeah, that laptop has a “full fledged non-Max-Q design” RTX 3080 Ti except it doesn’t. Yeah, it’s not the Max-Q variant, but it’s also not a “full fledged” part either.
Applying Hanlon’s Razor, I’d say LTT fell into stupdity due to Nvidia’s malice, though.
This is deceptive. Plain and simple. This is intended to confuse people. The problem is, this will also anger and frustrate people. But it doesn’t matter, Nvidia is getting their money. They have intelligent people working for them to figure this stuff out. This is intentional.
I have decided I will not reward this behavior any further. AMD is making pretty good progress catching up to Nvidia, and if they keep this up, my next build is very likely going to be AMD, not just because of performance and all that, but out of principle.
Yeah yeah they’re all terrible companies. As of now, AMD isn’t trying to implement deceptive naming schemes and are far more open source friendly.