The 5830 priced at $230-280 hardly makes sense when you can find 5850s at $300 and sometimes as low as $270
IF the card was priced at $200 then it wouldn't be a joke
NVIDIA GF100 & Friends speculation
- Thread starter Arty
- Start date
The 5830 priced at $230-280 hardly makes sense when you can find 5850s at $300 and sometimes as low as $270
IF the card was priced at $200 then it wouldn't be a joke
It's still GT200 (despite many adding a "b" at the end to distinct the 65nm from the 55nm shrink) though. If a reduced chip to 11SMs is named "GF104" then why isn't the 470 chip named something like "GF102" then?
And they've been never wrong before? Mark that I could be completely wrong too, but it's unlikely that any castrated chip carries an alternative codename. Under that reasoning the 5830 shouldn't be named Cypress but Begonia instead
It's still GT200 (despite many adding a "b" at the end to distinct the 65nm from the 55nm shrink) though. If a reduced chip to 11SMs is named "GF104" then why isn't the 470 chip named something like "GF102" then?
And they've been never wrong before? Mark that I could be completely wrong too, but it's unlikely that any castrated chip carries an alternative codename. Under that reasoning the 5830 shouldn't be named Cypress but Begonia instead
What? You clearly missed my point
Im not saying GF104 is the castrated part. Im saying that GTX460 can have the GF104 chip, and NOT a castrated GF100. Of course that a castrated chip doesnt have a different codename.
When I was talking about the GT200, I was refering to the xx5 part, like GTX285. In answer to the speculation that GTX460 and GTX465 use different chips (GF104 and GF100). That speculation doesnt seem feasible to me, as the xx5 moniker was used for different variations of the SAME chip. Understand now?
In answer to the speculation that GTX460 and GTX465 use different chips (GF104 and GF100). That speculation doesnt seem feasible to me, as the xx5 moniker was used for different variations of the SAME chip. Understand now?
Why not? It worked fine back when the 8800GTS' only differentiation for consumers was the amount of memory (512 versus 640) GT310 and GT315 also look like a "variation" but it's two completely different beasts.
Ok, as I said in the beginning of the argument "correct me if im wrong". I stand corrected then.
In CPU space it's common practice though.Mark that I could be completely wrong too, but it's unlikely that any castrated chip carries an alternative codename. Under that reasoning the 5830 shouldn't be named Cypress but Begonia instead
First of all, be very sure to tell you, GF104 silicon has been taped out.首先要很肯定的告诉大家,GF104流片成功了(好像地球人都知道了?);
性能方面NV对比的是HD5830,个别极端情况下(例如DX11的曲面细分测试)能达到HD5830 170%的性能(官方PDF里的数据);
NV将会在5月19日给AIC们提供GPU样品,听说Computex上会有用GF104演示的DEMO机,至于上市我估计最快也要6月底~7月了.nApoleon 发表于 2010-5-6 14:21
Performance NV contrast HD5830, individual extreme circumstances (such as DX11's curved segment test) to achieve the performance of HD5830 170% (official PDF in the data);
NV will be May 19 to AIC to provide GPU samples Computex heard there will be presentations with GF104 DEMO machine, as listed I guess the earliest by the end of June to July.
170% of HD5830?
Is Unigine Heaven counted as "extreme tesselation circumstances"?
Since 170% of HD5830 in Heaven would fit closer to HD5970 than HD5870
170% of HD5830?
Is Unigine Heaven counted as "extreme tesselation circumstances"?
Since 170% of HD5830 in Heaven would fit closer to HD5970 than HD5870
Well some people are calling GF104 Fermi II
Another guy some months ago, while i see him as a hoax, claimed GF104 was better then GF100, and not the mid range part. But almost all he said turn out BS so...
170% of HD5830?
Is Unigine Heaven counted as "extreme tesselation circumstances"?
Since 170% of HD5830 in Heaven would fit closer to HD5970 than HD5870
The marketing materials compare the GF104 and the 5830, the 170% is an extreme data point, with the same settings the GF100 does similar things to the 5870.
In same thread, comparing GF104 and RV870 die size, cfcnc:
ie Die size is roughly 10% larger than RV870 and the chip is presently crippled down to 5830 level so as not to adversely effect the GF100 above it.
The above is a bit rose colored glasses to say the least but anyway....
Yeah, doesn't sound very likely that nvidia wouldn't want to cannibalize the market of their 20% yield 530mm2 chip with the much cheaper to produce 350mm2/256bit one..
The 465 would still be good place to put all those chips that are capable of the said specs, but not 470 level
You got your wish! Just checked out newegg and theres a sapphire card going for $215 with free shipping, down to $199 if you're willing to deal with a mail in rebate! And some more cards can be had for 229-239 easily. Seems like there's been a bit of a price cut http://www.newegg.com/Product/Product.aspx?Item=N82E16814102878
Still on the wrong side of $200 for me to bite. Going from a 4850 1 gig to the 5830 isn't a big enough jump for me.
I think when the gf104 comes out things wil ldrop in price
Wait a minute... isn't that exactly what you wrote here 10 times before the GF100 release..?
Again, I certainly wouldn't hold my breath.
With a rumored price of 299$ for GF104 product (and in fact 180$ for GF106 which looks very uncompetitive to me) that seems rather unlikely...
If you keep saying it , hopefully it will come true !?!?! I mean if it isn't the gf104 then surely its ati's next card that will drop prices
Similar threads
- Deleted member 2197
- Architecture and Products
