安装 Steam
登录
|
语言
繁體中文（繁体中文）
日本語（日语）
한국어（韩语）
ไทย（泰语）
български（保加利亚语）
Čeština（捷克语）
Dansk（丹麦语）
Deutsch（德语）
English（英语）
Español-España（西班牙语 - 西班牙）
Español - Latinoamérica（西班牙语 - 拉丁美洲）
Ελληνικά（希腊语）
Français（法语）
Italiano（意大利语）
Bahasa Indonesia（印度尼西亚语）
Magyar（匈牙利语）
Nederlands（荷兰语）
Norsk（挪威语）
Polski（波兰语）
Português（葡萄牙语 - 葡萄牙）
Português-Brasil（葡萄牙语 - 巴西）
Română（罗马尼亚语）
Русский（俄语）
Suomi（芬兰语）
Svenska（瑞典语）
Türkçe（土耳其语）
Tiếng Việt（越南语）
Українська（乌克兰语）
报告翻译问题
讨论规则及指引
举报此帖
What makes you think the 5080 is going to be $1500? RTX 4080 Super is priced at $999 MSRP. I'd expect the RTX 5080 to be similar at $999-$1199. The RTX 5090 is a bit harder to predict with the memory bump but I'd still expect that to be in the ball park of $1699 - $1999. Also curious where you're getting that it is not as much of a jump as previous generations?
An article was saying that's the rumored price. Sure, it's a rumour, but knowing Nvidia, would it be that surprising? Especially if the 5090 is expected to break $2K-$3K price tag. They have no reason to keep the same prices as last gen.
They're using a new chip and apparently higher end components. That's just another excuse to justify a ridiculous jump in price. And it wouldn't be the first time they tried to oversell a lower tier. The original "4080 12 GB VRAM" was a cut down 4070(Ti) being sold as a 4080.
I don't think you know what the term "cut down" means. The "4080 (12GB)" wasn't a "cut down" 4070 Ti. It was the full AD104-400. The issue was that it wasn't the AD103-300 with less memory and the only obvious delineation in the branding was the memory difference; not that it was a "cut down" chip.
Was the article with the rumor just a trust me bro as their source? People said the same things prior to the 40-Super launches that they'd increase the pricing by $100 - $200; and the pricing was decreased.
The chip for the RTX 5080 is the GB203 and has a smaller die size than the AD103 in the RTX 4080 Super. Don't see that being a major driver in increasing the pricing $500.
No, the major driver would be that they can sell it for that price and there'll still be people who will buy it at that price. The point is that they don't care about the low end anymore, and are happy to even fleece the high end.
Anything to make the 90xx series look like a better deal and value than ANY of the cards lower in the stack. Although even enthusiasts would have to raise an eyebrow here, since having a budget of $1600 before doesn't mean you now suddenly have a budget of $2-$3K. They might as well be AI and workstation cards. But Nvidia won't call them that, in hopes that "gamers" will buy them.
Have you seen some of the new intel arcs coming out that rival the 4060/4070. I had my doubts at first but intel seems like they are entering the field on a good note.
The 4080 launched at 1200. That was a big issue at the time since the previous 3 gens launched between 600-800. But they learned their lesson and dropped the 4080 super for 999. And the super was way more successful. NVIDIA knows they will make more money if they keep it around that price. And considering how powerful the 5080 is gonna be, 1200 would absolutely be reasonably priced. And it would be fairly consistent since the 1080. The only anomaly was the 3080 when it launched at a lower cost than the 2080. But other than that, each xx80 card has seen about an increase of 200 each launch. And again, it would be reasonable considering how powerful it is.
NVIDIA is not the enemy here. That would be the scalpers. But they will most likely target the 5090 this time around. And of course the tariffs coming to the USA (if you live there) that might increase it on the AIB market.
My guess is they are gonna launch the 5080 at 999-1199. Probably 1099. And the 5090 at 1799
They do look like they've made some significant progress with Battlemage. Hopefully they didn't kill off the big Battlemage chip and they still plan on launching an ARC B770 / B780 around the $400-500 price point to compete at the mid-high tier with good value. It looks like they've gotten on the right track software wise and I'd credit Intel more than anything in regards to AMD's decision to forego the enthusiast market and focus more on the mid range. I think AMD realized they were about to be squeezed out of the market from both ends.
I don't think the RTX 5080 is going to be a massive uplift from the 4080. I would expect it to be a similar uplift from the 3080 to the 4080 (without FG). The RTX 5090 however is likely going to be an even larger gap this time.
The RTX 4090 uses the AD102-300-A1 GPU which is a cut down version of the full AD102-400 with 16384 CUDA cores.
The RTX 5090 will use the full GB202-400-A1 from the leaked bench marks with 24576 CUDA cores.
Which you can't directly compare different generational cores directly; a newer architecture core-to-core is typically also an improvement in performance. So there will likely be a substantial increase from the 4090 to the 5090.
The RTX 4080 Super uses the full AD103-400-A1 GPU with 10240 CUDA cores.
Whereas the RTX 5080 is expected to use the full GB203-400-A1 GPU with 10752 CUDA cores.
EDIT: I should also note that even if those leaked benchmarks weren't accurate and the production for the 5090 will be a cut down GB202-300-A1 that will still end up being 21760 CUDA