Graphics Cards

Nvidia super: everything we know so far

Table of contents:

Anonim

In this article we will talk about everything you need to know about possible new Nvidia SUPER graphics cards. And it is that the preparations prior to the conferences in the great fairs, in this case the Computex and the E3, always nourish us with succulent leaks and leaks.

In this anniversary year for AMD in which it celebrates its Resurrection Half a century old, the company has prepared for us a non-stop launching spread over the 12 months.

Apisak, pre-filtering the Ryzen 3950X. It also leaked (confirmed) at E3.

At Computex were the new Ryzen, and at E3 the graphic novelties with RDNA and Navi.

Both of them were cannon fodder from the leaks and we were able to find out practically all the details before the presentation took place at the scheduled time and day.

In any case, let's focus on the fact that approximately one week before the celebration of the first of the two events, Nvidia moved the file by publishing a teaser… and that each one understood it in the way they liked best. For effects, an autoleak (lifelong self-shield).

Index of contents

The fact is that I left it there, in advance. SUPER was running.

The fairs were held and we knew absolutely nothing about the subject, at least officially from Nvidia. Many concluded that Nvidia had teased us.

What we did know, everyone, were the tricks that AMD reserved for itself since it raised its cards. In keeping with expectations, its new GPUs are 'progressing well', but taking its previous 'achievements' as a reference. No danger at the moment.

AMD Radeon RX Series Specification Comparison
AMD Radeon RX 5700 XT AMD Radeon RX 5700 AMD Radeon RX 590 AMD Radeon RX 570
Stream Processors 2560

(40 CUs)

2304

(36 CUs)

2304

(36 CUs)

2048

(32 CUs)

Texture Units 160 144 144 128
ROPs 64 64 32 32
Base Clock 1605MHz 1465MHz 1469MHz 1168MHz
Game clock 1755MHz 1625MHz N / A N / A
Boost Clock 1905MHz 1725MHz 1545MHz 1244MHz
Throughput (FP32) 9.75 TFLOPs 7.9 TFLOPs 7.1 TFLOPs 5.1 TFLOPs
Memory clock 14 Gbps GDDR6 14 Gbps GDDR6 8 Gbps GDDR5 7 Gbps GDDR5
Memory Bus Width 256-bit 256-bit 256-bit 256-bit
VRAM 8GB 8GB 8GB 4GB
Transistor Count 10.3B 10.3B 5.7B 5.7B
Typical Board Power 225W 180W 225W 150W
Manufacturing Process TSMC 7nm TSMC 7nm GloFo / Samsung 12nm GloFo 14nm
Architecture RDNA (1) RDNA (1) GCN 4 GCN 4
GPU Navi 10 Navi 10 Polaris 30 Polaris 10
Launch Date 07/07/2019 07/07/2019 11/15/2018 04/08/2016
Launch Price $ 449 $ 379 $ 279 $ 179

This results in the following:

  • Products with specifications within its premise 1.25 / 1.5.

Don't waste time dreaming about RTG products outside of the current 'HE Operational Range'.

  • Directly related to the above, competing in the mid and high-mid segments while waiting for time and their commitment to improvement and scalability (in the hands of those who manufacture their designs) allow them to go further.
  • Mark prices following the logic of the market and its laws (not written by the way).

Now the interpretations

Given what was seen in the E3 launch, technological advances and improvements in AMD's RTG division continue to be far from Nvidia's performance and efficiency beyond the 2070 rung.

Even currently offering competition within the range they have been able to achieve, they do so with less efficiency, in any case this is not a problem in itself.

Likewise, we will recognize that neither is it that some use a lower node (7nm little mature) and that Nvidia continues without disheveled (and voluntarily) in one that a priori penalizes it (12nm very mature and optimized).

Everything that points to 2080 or 2080Ti is left out, but let's return to what the article tries to focus on.

AMD has positioned its new competitive products with a decrease in the price of what they understand as "equivalents" of Nvidia.

If anyone expected them to 'squeeze' and try to win 'the game' by pulling prices, they were wrong. Unfortunately this does not work like this, unfortunately for the consumer.

In commerce it is considered a mistake to be cheaper than necessary, and much more 'first'. To compete (with what you have), it will be valued to be cheaper than with whom you try to show yourself as an alternative. And being cheaper is costing 1 hard less than the other.

On the one hand, the one who wants a Radeon will surely buy a Radeon (unless it is more expensive than the Geforce equivalent) and on the other, the one who is not an expert and compares apparent (basic) technical characteristics goes to the peels'.

Taking all of the above into account, let's try to predict the materialization that SUPER will carry out in a couple of days (and for which we already have leaks via wccftech).

SEGMENTATION GPU MSRP
NVIDIA GeForce RTX 2080 Ti TU102A $ 1199.99
NVIDIA GeForce RTX 2080 Ti TU102 $ 999.99
NVIDIA Geforce RTX 2080 SUPER TU104-450 $ 799.99
NVIDIA GeForce RTX 2080 TU104A-400 $ 799.99
NVIDIA GeForce RTX 2080 TU104-400 $ 699.99
NVIDIA GeForce RTX 2070 SUPER TU104-410 $ 599.99
NVIDIA GeForce RTX 2070 TU106A-400 $ 599.99
NVIDIA GeForce RTX 2070 TU106-400 $ 499.99
NVIDIA GeForce RTX 2060 SUPER TU106-410 $ 429.99
NVIDIA GeForce RTX 2060 TU106A-400 $ 349.99

Nvidia should not have felt 'threatened' by RTG's Computex / E3 products.

They did not need to link absolutely anything during the fairs to their previous notice in the form of a Teaser without content. With all the calm in the world, once they have finished, it seems that they are simply going to tweak us on the GDDR6 and casually adjust prices.

WE RECOMMEND AMD RX 5600 XT has not achieved the best launch

The hardware they may have prepared ( porsiaca ) with specifications higher than those currently marketed, they keep on keeping. Yes, it is another mistake to try to enlarge your advantage over the competition when you are already ahead, it is unnecessary.

At least under normal market conditions, there are exceptions as we have been seeing for a few years in the case of CPUs between Intel and AMD, but in this case there is an explanation: Sales are not prioritized, but the marketshare.

Consequently, Nvidia's mysterious (and versatile) Teaser SUPER will ultimately mean immediately repositioning the products it considers affected by the movement of its competitor.

Due to the leaks, the only improvement in the new SUPER variants of the Geforce would be the use of memories programmed at higher frequencies.

Something that in principle we can think that Nvidia can do at any time, but has decided to do, casually, now.

In this way it will again be ahead (and we have already commented that it does not need to be much, simply ahead) in performance.

At the same time you can play with the prices, adjusting the versions that the competition has used to 'reach' and put them minimally below (and go back to those of ' the skin is the skin' ) and replace with the new SUPER variants while maintaining old prices. (And go back to those who compare the specs superficially.)

The conclusion we can reach is that, currently, GPU prices are set by Nvidia as it dominates the market.

The competitor (AMD), who is aware that at the moment he can hardly choose to take the sales that the marketing and the blind predilection of customers for their color, brand or whatever they represent, in addition to those that depend on the rules that previously I have commented that they rule in the market (price to the peel, or performance looked at above), it is dedicated to act strictly based on this using the other's price as a starting point.

I wonder if it really is beneficial for the consumer (and their pocket) that Nvidia launches the SUPER or if it would have been enough to have done absolutely nothing and leave the SUPER teaser as an anecdote after having verified what AMD currently has.

We recommend reading the best graphics cards on the market

Surely we will not know until the Radeon 5xxx, Geforce RTX and SUPER can be faced on the test bench and their performance and value in relation to each other are confirmed.

Graphics Cards

Editor's choice

Back to top button