MD unveiled its new Radeon RX 5700 line of graphics cards with 7nm chips at E3 last month, and with just days to go before they launch on July 7th, the company has announced new pricing. In the "spirit" of competition that it says is "heating up" in the graphics market -- specifically NVIDIA's "Super" new RTX cards -- all three versions of the graphics card will be cheaper than we thought.
The standard Radeon RX 5700 with 36 compute units and speeds of up to 1.7GHz was originally announced at $379, but will instead hit shelves at $349 -- the same price as NVIDIA's RTX 2060. The 5700 XT card that brings 40 compute units and up to 1.9GHz speed will be $50 cheaper than expected, launching at $399. The same goes for the 50th Anniversary with a slightly higher boost speed and stylish gold trim that will cost $449 instead of $499.
That's enough to keep them both cheaper than the $499 RTX 2070 Super -- we'll have to wait for the performance reviews to find out if it's enough to make sure they're still relevant.
AMD fires back at 'Super' NVIDIA with Radeon RX 5700 price cuts
High refresh rates really aren't an issue unless you play a certain old game which requires 240hz, and adaptive sync technologies have helped mitigate the need to keep 60/75 FPS at all times, but when a 2080 Ti can't even achieve that across the board...Considering the likely development cycles it will probably be 2022 before uncompromised 4k60 gaming at the HIGH END becomes a reality, with it trickling down to the mid-range in 2025. nVidia will no doubt achieve this first with Intel probably second as they have the money to spend on R&D and Intel is going to want to come in like a hurricane, but the question is how fast AMD will adapt to a three manufacturer market. Next year's cards are going to be a bust by all early accounts. It's going to be Navi based, not next gen, so quite likely it's going to be pushing 250-300w if it is an "nVidia killer" UNLESS Navi has a glaring inefficiency in it they can rectify since we know that AMD will be implementing their hybrid ray tracing technology. Then there's the cost of HBM2E, which we know those cards will use, which will add a large cost to the cards, to Vega it added $160 plus the cost of the packaging...Plus you figure that if the mid range cards are costing $400-$500, AMD's high end cards will be $750-1000 like nViida...
Next year is going to be very interesting for sure. nVidia's releasing their RTX 3000 series which will likely be a refinement of the RTX 2000 series so I think we can expect 25% higher performance with a larger jump with ray tracing, Intel's going to release their new cards next year starting at $200 and will likely target the entire mid range, and, let us not forget, if AMD will be releasing their 4000 series processors which should hopefully fix the 3000 series' shortcomings and will be the final processors for Socket AM4, and Intel will release their long awaited 10nm units.
pokester wrote:
you are absolutely correct and the Super cards really rectify those shortcomings and up the performance to a point that starts to justify upgrading from GTX cards.
I consider my GTX 1060 to be an example of 16 nm hardware but recent marketing nonsense makes that metric dubious
I am starting to look at number of transistors per square millimeter and maybe a better metric. Simple calculation will clarify the real density of the logic.
so with 4400M transistors into a 200mm chunk of silicon, i get 22M per square mm
The 3000 series is already 7nm, but TSMC has two variants that the 4000 series could be based on. 5nm will also be released next year so it is -possible- AMD will use it to stay ahead of Intel's process, but one would imagine it would be much more costly to redesign for that node especially when a noticeable performance gain will be attained simply from perfecting yields to eliminate the "one fast many slow" arrangement they use now, plus holding off on 5nm would give TSMC time to perfect the node for the new architecture 5000 series so AMD will not have the issues with them they do now with the 3000 series. Also the fact that 5nm isn't going to ramp up until the middle of next year would put AMD behind their "7" launch strategy, and would also delay the 5000 series as well.
Nvidia seem to have responded / pre-empted a few more things versus AMD RX5700XT with their latest Driver update.
(1). They have added a new Sharpening Filter and added the number of games supported by Nvidia Freestyle. That seems to be a response to AMD Contrast Adaptive Sharpeneing (Radeon Image Sharpening). I just took a look at it. The new Sharpening menu appears above the SpecialFX one in the Game Filter. They have added support for many more games. Use of ReShade is blocked on CSGO, so I am unable to use CAS.fx or Smart_Sharp.fx or even Software based Ray Tracing on CSGO with my AMD GPUs.
It provides two slider controls. Sharpen, and Ignore Film Grain.
Here are a couple of screenshots showing the filter in action:
Default:
New Sharpening:
(2) They have added Integer Scaling Support in Control Panel.
It looked like AMD was going to be first to deliver Integer Scaling - it is in the "Vote for new features List" for Adrenalin 2019.
Intel implemented it for their new GPUs and now Nvidia have it.
I do not know if integer scaling is implented in the RX5700XT.
If I read the press release correctly, the new sharpening filter is an addition to nVidia Freestyle and not a Turing specific feature, which means that any currently supported nVidia card, from the 600 series to the new RTX series, should be able to use it in any of the currently supported 600 games, which is in contrast to AMD's requirement of a Navi card although RIS works in any game. That's a large win for nVidia as AMD has no way of delivering an image sharpening filter to Vega and earlier products.
The new sharpening filter does throw a wrench into things. AMD's higher end Polaris card, the RX 590, is now $190-$200, and the GTX 1660 can be found for $215-$220. The 1660 already consumes nearly half as much power, and is much quieter, than the 590, which already are two things very much in favor for nVidia considering the very small price difference, but add in image sharpening and it becomes very difficult to recommend using an AMD card even for the most popular resolution, 1920x1080.
