Tech blogs and sites have been aware that NVIDIA's RTX cards are affected by a "class system" since these GPUs started hitting the market. This system stems from the existence of 2 variants of the same Turing (TU104 or TU106) chips.
For example, there is such a thing as a TU104-400-A1 chip - and also such a thing as a TU104-400A-A1 chip. The first "A" in this nomenclature is what differentiates the two. These variants tend to be found in higher-end products that can be overclocked, whereas the "non-As" are often in lower-end cards that are not rated for the same.
For example, the TU104-400-A1 chip was found to be incorporated in the basic ASUS RTX 2080 Turbo card, whereas the TU104-400A-A1 featured in its ROG Strix counterpart. Then again, this is not necessarily a hard and fast rule: a kind of OC-mode chip lottery is also possibly in effect when buying RTX cards thanks to this dichotomy.
A contributor to TomsHardware.de claims to have exclusive information on new variants of Turing GPUs. They are apparently to be called the TU104-410 & TU106-410 series of chips - and that's it: no "A-class" sub-variants here.
The correspondent (Igor Wallossek) also now projects that overclockability will be constant across these new series. Therefore, a shift to these new dies among manufacturers may render more effective and flexible RTX performance more affordable in the future.
Are you a techie who knows how to write? Then join our Team! Wanted:
- Specialist News Writer
- Magazine Writer
- Translator (DE<->EN)
Details here
Source(s)
Join our Support Satisfaction Survey 2023: We want to hear about your experiences!
Participate here
Top 10 Laptops
Multimedia, Budget Multimedia, Gaming, Budget Gaming, Lightweight Gaming, Business, Budget Office, Workstation, Subnotebooks, Ultrabooks, Chromebooks
under 300 USD/Euros, under 500 USD/Euros, 1,000 USD/Euros, for University Students, Best Displays
Top 10 Smartphones
Smartphones, Phablets, ≤6-inch, Camera Smartphones