• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Question 1660 Ti details

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Your Vega64 and 580 are both one spot too low and your 2070 and 2060 are both 3 spots too high. Many of your prices seem off as well, at least for the US (currency exchange included).
They're aggregates, based here. They're relative and wouldn't show which ones are better at 1080p, 4K gaming etc.

The prices are the cheapest ones I could find on a combination of Overclockers UK, Ebuyer, Scan, Box and Amazon. They're accurate.
 
They're aggregates, based here. They're relative and wouldn't show which ones are better at 1080p, 4K gaming etc.

The prices are the cheapest ones I could find on a combination of Overclockers UK, Ebuyer, Scan, Box and Amazon. They're accurate.

You can't judge videocard gaming performance from one benchmark, let alone passmark. You should at minimum use a site that does composite performance charts across a good selection of games, such as:

https://www.computerbase.de/2019-02/geforce-gtx-1660-ti-test/3/#diagramm-performancerating-2560-1440
zSQrr2O.png


Pricing is also tricky as just looking for the lowest price can lead you to very limited time offers (such as when a Vega56 model was being sold for $250 in the US for like a day) that the vast majority of people won't be able to take advantage of. This is also market dependent which is why I gave the caveat of US prices.
 
You can't judge videocard gaming performance from one benchmark, let alone passmark. You should at minimum use a site that does composite performance charts across a good selection of games, such as:

https://www.computerbase.de/2019-02/geforce-gtx-1660-ti-test/3/#diagramm-performancerating-2560-1440
zSQrr2O.png


Pricing is also tricky as just looking for the lowest price can lead you to very limited time offers (such as when a Vega56 model was being sold for $250 in the US for like a day) that the vast majority of people won't be able to take advantage of. This is also market dependent which is why I gave the caveat of US prices.
The above list is mostly accurate.

I, of course, didn't include anomaly prices or prices on special offer. The prices listed are the lowest general prices from very large online stores.

I stand by the list.
 
Interesting - so the first rumours had tensor cores, then the release card doesn't but has got FP16, and it turns out that tensor cores are what give the 2xxx cards their FP16 compute power (as used by Wolfenstein for example).

So what's the betting that F16 cores on the 1660Ti are in fact tensor cores, but Nvidia's decided to not call them that? Perhaps there's not enough to do a good enough job on DLSS, or perhaps they don't have enough super computer compute time to generate DLSS algorithms for the 16XX cards?
 
Interesting - so the first rumours had tensor cores, then the release card doesn't but has got FP16, and it turns out that tensor cores are what give the 2xxx cards their FP16 compute power (as used by Wolfenstein for example).

So what's the betting that F16 cores on the 1660Ti are in fact tensor cores, but Nvidia's decided to not call them that? Perhaps there's not enough to do a good enough job on DLSS, or perhaps they don't have enough super computer compute time to generate DLSS algorithms for the 16XX cards?
They are apparently brand new dedicated FP16 cores and this is the first card with them. These FP16 cores are much smaller than Tensor cores, so there's little doubt.

From the Anandtech piece on TU116.
 
Back
Top