• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

GamerNerves

Limp Gawd
2FA
Joined
Dec 23, 2023
Messages
129
A thread for questions, news, reviews, impressions, comments and opinions regarding RTX 5070 (12 GB).

Here is my question in the spoiler:
How popular the RTX 5070 will be at launch? Will it have immediate availability issues? Is the initial stock much larger than the models' which were already released (5090, 5080, 5070 Ti)?
 
Last edited:
I'm sorry, but I couldn't find a truly fitting topic for my next question, so I made this new thread for general questions, news, comments and opinions regarding RTX 5000 series (Blackwell).

My question is in the spoiler:
How popular the cards RTX 5070 Ti and 5070 will be at launch? Will both or either one have immediate availability issues? Is the initial stock much larger than the models' which were already released (RTX 5090, 5080)?
Nobody knows for sure, if I were to guess the stock will be high since they will be competing directly with AMD's launch which is three weeks later.
 
no one knows, yet.
title should say 5070s, otherwise this looks like every other 5000 series thread....
 
Quoting Lakados from another thread:

The big thing I wonder about is this, will the arrival of the 5070TI's be accompanied by new shipments of the 5080's? They use the same chip but the 5070TI is cut down, so theoretically if the yields of the GB203 are good there should be fewer 5070TI's produced than 5080's at any given time.
It also means that the 5080 could have it's price reduced to that of the current 5070TI should AMD push them hard because the costs from one to the other are virtually identical, the 5080 needs a few extra RAM chips, but the 5070TI needs extra fab time to cut off the bad sections, the two costs more or less balance out, and I doubt that the AIB's would be using different boards for the two chips, it would cost them more to build a slightly cut down PCB for the 5070TI than it would to just increase production of the PCB used on the 5080 just due to economy of scale.

He brings up a good point. The 5070ti will be slower, but performance to price ratio, it may actually be better (assume $800 when prices settles vs $1050-1100+). Makes little sense to make a lot of those if they can be turned into 5080s with higher profit margins.

RTX 5070, I suppose we'll have to wait and see but I think the 5070ti will be a popular model given its supposed price.
 
Having to put a bet, yes in the first 48 hours of their launch they will be hard to buy with a couple of click on bestbuy, nvidia, amazon and elsewhere.

FOr sure for the upcoming 5070ti, 5070 is harder to say, no release date yet, lot could change between now and then, 263mm gpu volume could be quite different, AMD could have restarted selling gpu in between.... and China new year finish tomorrow, in a couple of weeks some things could have "restarted"
 
Based on the 5080 and 5090 launch, I would guess that there is little supply at the start and Nvidia counting on triggering FOMO. Disappointed with the 5080 and not sure if the 5070 cards will be all that great. There could of course be very poor yields on perfect GB203, but many almost fully functioning chips, which could lead to lots of 5070 Ti cards.

The 5080 is priced more like the MSRP was $1250-1300, rather than $999 and I wouldn't be surprised if the same happens to the 5070 cards (20-30% above MSRP for the AIB cards). I'll wait until AMD has launched the 9070 XT and 9070 and are available at the stores before I make any decisions.
 
There could of course be very poor yields on perfect GB203,
They are mid-size on a very mature node, should be similar to the 4080super yield by the end of their production, using yield calculator do show 60-80% type of yield for the same size of die for 0.06-0.12 error per cm type of node, not sure how perfect those are.

The almost perfect die but not perfect could be going into the 5090 mobile chips (10,496 out of 10,752 enabled core) , if TPU is right:
https://www.techpowerup.com/gpu-spe...03 GPU uses the,Ultimate (Feature Level 12_2).
 
Microcenter RTX 5070ti pricing. Not sure if these are price holders, but looks like a massive increase over MSRP like we see for higher end cards. Only MSRP model is the PNY. Even the Gigabyte Eagle is $900, and those are typically one tier above the Windforce. If it is $900, I don't see Gigabyte releasing a model sub $800.

Really wish I bought an RTX 4070ti Super last year for the 16GB of VRAM...
 
I'll get in trouble talking about this...but I couldn't wait

Embargo doesn't lift on the 5070Ti until the 19th for MSRP models and the 20th for the non-MSRP models, but there is a serious issue here with the pricing and what we are being told...so I need to discuss this NOW!...


View: https://www.youtube.com/watch?v=LgAb5bmcTjk
 
  • Like
Reactions: jaffy
like this
The least surprising result imaginable.

+20% the 4070, about a 4070super has pixel counting the Nvidia slide at announcement or the specs predicted.

Looking at 9070-9070xt offer a 5070super replacement could be needed relatively quick, they left good room for it.
 
Makes me wonder what else nVidia is lying about?
?

5070 +20% the 4070, about a 4070super was "clear" enough for an enterprise level customer to discern here, 100% of hardforum message board users got that from Nvidia presentation. Not a single person in a single AIB thought when they bought 5070 die thought they were getting 4090 performance to build their gpus.

The Microsoft-google-amazon-oracle make their own test (benchmark for what they do will not exist or be relevant quite often) and their teams know things more than marketing slides. And the presentation was clear enough that the 4090 performance talk was :
and the reason
for that is because we're generating
most of the pixels using pixels using
our tensor cores so we retrace only the
pixels we need and we generate using
artificial intelligence all the other
pixels we have as a result the amount of
 
Last edited:
I'm just mad no one sold me their 4090 after the CES announcement. :p


They don't want to sell gaming cards, the 5000 series is just keeping their foot in the door incase they need the business in the future.
 
?

5070 +20% the 4070, about a 4070super was "clear" enough for an enterprise level customer to discern here, 100% of hardforum message board users got that from Nvidia presentation. Not a single person in a single AIB thought when they bought 5070 die they were getting 4090 performance to build their gpus.

The Microsoft-google-amazon-oracle make their own test (benchmark for what they do will not exist or be relevant quite often) and their teams know things more than marketing slides. And the presentation was clear enough that the 4090 performance talk was :
and the reason
for that is because we're generating
most of the pixels using pixels using
our tensor cores so we retrace only the
pixels we need and we generate using
artificial intelligence all the other
pixels we have as a result the amount of
Let’s start with the 5070 does not equal 4090 performance.

At best the marketing and slides were extremely misleading to consumers, at worst were blatant lies.

Why does nvidia continue to step on its own feet? In this day and age with real time testing, tech reviews and YouTube, you’re a fool to try to pull wool over people’s eyes. Nvidia needs to wisen up.
 
Yes, extremely misleading to the average consumers, world of some many free reviews with good reach around that at least it get corrected for anyone that care (more people watch review than CES presentation would be my guess, about no one watch those, long one hour+ mostly about enterprise product) or go on Nvidia.com using a browser.

They try to mislead softbelly tech "news" ecosystem (that is giant) that will go and mislead consumer.

Why does nvidia continue to step on its own feet?
because they do not have the performance numbers would be my guess, the 4090 had them and still used those strange benchmark, but the 4070 did not had nice results, if you do not want to break the format down the stack, it is not an easy to sell and get people excited about refresh.

AMD will succeed to push them into better card on the upcoming super revision and they will not take too long to make it, 3GB gddr7 becoming available would be a perfect excuse to do it without having to admit any fault in your previous launch perf/$ stack. 5060-5060 ti pricing could be the start of that pressure, if it is there and will be a clue.
 
Essentially an overclocked RTX 4070 Super. Lame. Doesn't even have improved ray tracing performance in any meaningful way, and still only has 12GB of VRAM.
 
This card is getting trashed in every review. Nvidia went too far with the “4090 performance” shit. How did they think that was a good idea? Do they think nerds are that dumb?
 
This post is in many parts what I have recently wrote on this forum, so it's repetative:
This card is getting trashed in every review. Nvidia went too far with the “4090 performance” shit. How did they think that was a good idea? Do they think nerds are that dumb?
This is the perfect time to witness if reviews actually have meaningful impact in swaying gamers to the other option, Radeon, or alternatively make people avoid RTX 5070 before a price cut. Of course the price will be cut if they stand on shelves and by that a Super variant will come much sooner to fill the big price gap between 5070 and 5070 Ti. I don't personally see RTX 5070 as a negative release as many of the outlets, because it all depends on where are you coming from (I have no dGPU at all!), but indeed it barely moves the needle forward in the price class, but Gamer's Nexus had some results where in practice it is on par with either RTX 4070 Ti or RX 7900 XT (game dependent). Once the new consoles come out, even without the Radeons existing at all in the desktop market, Nvidia has to offer some improvements in value to keep the PC audience occupied and the PC platform the leading one visually, which is their field, and for the sentiment in the first place to be a PC gamer other than the PC specific titles, so a stopgap generation fits their situation quite well, when they can boust in the next generation already with some gains perhaps.

Is AMD moving the needle a thing to the wider audience of potential shoppers in this price class? Of course the fact itself means nothing, the moaning we see is purely from hobbyists and enthusiasts, because to the regular customer, once again, applies the following:

  • Buys under the influence of market presence; in practice, what is offered to him, what is available in stores.
  • Might believe that features he sees inside games, in marketing or elsewhere are important.
  • Asks a friend on what to buy, or buys the same, or asks online or reads others' posts online.
  • Wishes to make a secure purchase, since troubleshooting may be out of his ability; accordingly for him may be recommened Geforce.
  • Does something on his computer that requires CUDA or other Geforce feature that he is aware of.
  • Considers resalability.

I have built Radeon systems for my friends, but this is because I can help them if need be when issues happen. For others, at least today, I rather recommend Geforce to avoid any possible hassle, like the Windows update replacing your driver on Radeon, which borks the software suite. Reviews about any product categories are viewed, so the regular gamer likely checks some GPU reviews too online, but I don't think bashing Geforce ensues anything other than slight confusion. Anybody, non-tech people too, can understand from them that the enthusiast sentiment is against Geforce, but it only spells the feelings around the products, since in the conclusion of any proper review there are the value charts, which are favorable for the 5070. The Radeon reviews will affect this aspect of course, but do they make 5070 clearly a bad purchase, even if you get something like 20 % less performance for the same money? No, of course not, the product does what it is supposed to, even if it's not as valuable from the perspective of rasterization performance. The situation still simply is that only thing Radeon offers over Geforce, is extra VRAM and rasterization performance - that's about it. I'm all for the VRAM, but it would take effort to present the benefits of it for the pondering buyer, since usually reviews only speculate on the need of it.

My aim is not to defend the RTX 5070, since it represents stagnation quite strikingly, but there are no guilty parties here unless you believe that the duopoly agrees upon things together. Neither party is wishing to make major value products, the other suffering from low trust factor and is behind in some other aspects, while the other has no incentive to offer anything better. Since progress is slower, games are fitted to the hardware people cling on, so games continue to run in any case, and if 8 GB cards are still to be sold widely in this generation, what the RTX 5060 might have, then this amount of VRAM is taken into account for longer. I always wish that higher amounts of VRAM is taken into account also, but what I mean is, there is no need to worry about anything or complain too much, because the customer base defines what they are wishing to pay for, which is likely not an RTX 5070 for the majority. Also, I'm a bit surprised that so little kudos are given to Nvidia for offering DLSS upscaling in it's latest iteration all the way down to the RTX 2000 owners. In the case of these customers it can extend the lifespan of the product, and someone who is on, let's say RTX 2070 or 2070 Super, which both have 8 GBs of VRAM, 8 GBs being taken into consideration further is a happy thing for them.

Stagnation to those hungry for advancements is a negative thing surely, but it is not all negative - there's not much reason to complain in the end, since nobody forces you to upgrade. It hurts me too to pay for RTX 5070, since I'm quite set on that, but if it just works and games run adequately, I'm sat. :asmile:

EDIT. grammar
 
Last edited:
This post is in many parts what I have recently wrote on this forum, so it's repetative:

This is the perfect time to witness if reviews actually have meaningful impact in swaying gamers to the other option, Radeon, or alternatively make people avoid RTX 5070 before a price cut. Of course the price will be cut if they stand on shelves and by that a Super variant will come much sooner to fill the big price gap between 5070 and 5070 Ti. I don't personally see RTX 5070 as a negative release as many of the outlets, because it all depends on where are you coming from (I have no dGPU at all!), but indeed it barely moves the needle forward in the price class, but Gamer's Nexus had some results where in practice it is on par with either RTX 4070 Ti or RX 7900 XT (game dependent). Once the new consoles come out, even without the Radeons existing at all in the desktop market, Nvidia has to offer some improvements in value to keep the PC audience occupied and the PC platform the leading one visually, which is their field, and for the sentiment in the first place to be a PC gamer other than the PC specific titles, so a stopgap generation fits their situation quite well, when they can boust in the next generation already with some gains perhaps.

Is AMD moving the needle a thing to the wider audience of potential shoppers in this price class? Of course the fact itself means nothing, the moaning we see is purely from hobbyists and enthusiasts, because to the regular customer, once again, applies the following:

  • Buys under the influence of market presence; in practice, what is offered to him, what is available in stores.
  • Might believe that features he sees inside games, in marketing or elsewhere are important.
  • Asks a friend on what to buy, or buys the same, or asks online or reads others' posts online.
  • Wishes to make a secure purchase, since troubleshooting may be out of his ability; accordingly for him may be recommened Geforce.
  • Does something on his computer that requires CUDA or other Geforce feature that he is aware of.
  • Considers resalability.

I have built Radeon systems for my friends, but this is because I can help them if need be when issues happen. For others, at least today, I rather recommend Geforce to avoid any possible hassle, like the Windows update replacing your driver on Radeon, which borks the software suite. Reviews about any product categories are viewed, so the regular gamer likely checks some GPU reviews too online, but I don't think bashing Geforce ensues anything other than slight confusion. Anybody, non-tech people too, can understand from them that the enthusiast sentiment is against Geforce, but it only spells the feelings around the products, since in the conclusion of any proper review there are the value charts, which are favorable for the 5070. The Radeon reviews will affect this aspect of course, but do they make 5070 clearly a bad purchase, even if you get something like 20 % less performance for the same money? No, of course not, the product does what it is supposed to, even if it's not as valuable from the perspective of rasterization performance. The situation still simply is that only thing Radeon offers over Geforce, is extra VRAM and rasterization performance - that's about it. I'm all for the VRAM, but it would take effort to present the benefits of it for the pondering buyer, since usually reviews only speculate on the need of it.

My aim is not to defend the RTX 5070, since it represents stagnation quite strikingly, but there are no guilty parties here unless you believe that the duopoly agrees together on all of this. Neither party is wishing to make major value products, the other suffering from low trust factor and is behind in some other aspects, while the other has no incentive to offer anything better. Since progress is slower, games are fitted to the hardware people cling on, so games continue to run in any case, and if 8 GB cards are still to be sold widely in this generation, what the RTX 5060 might have, then this amount of VRAM is taken into account for longer. I always wish that higher amounts of VRAM is taken into account also, but what I mean is, there is no need to worry about anything or complain too much, because the customer base defines what they are wishing to pay for, which is likely not an RTX 5070 for the majority. Also, I'm a bit surprised that so little kudos are given to Nvidia for offering DLSS upscaling in it's latest iteration all the way down to the RTX 2000 owners. In the case of these customers it can extend the lifespan of the product, and someone who is on, let's say RTX 2070 or 2070 Super, which both have 8 GBs of VRAM, 8 GBs being taken into consideration further is a happy thing for them.

Stagnation to those hungry for advancements is a negative thing surely, but it is not all negative - there's not much reason to complain in the end, since nobody forces you to upgrade. It hurts me too to pay for RTX 5070, since I'm quite set on that, but if it just works and games run adequately, I'm sat. :asmile:
Wat?
 
Launch day. Microcenter Sharonville OH shows 2 MSRP cards so far. No idea how many cards they have though only 6 skus on the site as of now.

IMG_6823.jpeg
IMG_6824.jpeg
 
Another typical Founder's debacle. I was watching Bestbuy and it went from "Coming Soon" to "Sold Out" without ever hitting "Add to Cart."
 
Store opened at 9AM the 2 skus at MSRP sold out in 20 minutes. ..

As of now 9:30AM 4 skus showing limited stock at 599 and up.

Edit: Micro center Sharonville OH
 
And 1 hour in the 599 models are gone. Will people go for the last 2 skus at 699 and 739!? For a 5070? Madness.

Micro Center Sharonville OH
 
And I sure as hell wasn't paying $650 for a 5070 without at least looking to see what 9070 availability looks like.
Yeah 549-600 is the limit for a 5070 IMO. It’s a 4070 super+ with a small or no price movement, frame gen and drops old physX if people care. It’s an OK card for that money for the green team. Paying more just buy up to a stronger card with more vram etc.
 
Last one standing. Silly money for a 5070 but here we are. I’ll stop spamming the thread now..😎


IMG_6825.jpeg
 
From a 1080 Ti to a 3070 and a 3080, I don't see the RTX 5070 as much of an upgrade over the RTX 3070. The RTX 3070 is a 4 and a 1/2 year old card, and the RTX 5070 is only around 32% - 35% faster in 1080P and 1440P. The same time gap between the GTX 1070 and RTX 3070 was 4 and 1/2 years, but the difference in performance was up to 95% in 1080P. Paying more, but getting less, and newer games sucking more... I think I'm in the "Let's wait and see what the next gen has to offer" camp.
 
Last one standing. Silly money for a 5070 but here we are. I’ll stop spamming the thread now..😎


View attachment 714756

My brother used to live in Dayton, so I'd occasionally get out to the Sharonville MC. Right now, the closest one is 125-ish miles away from me, and I'm not driving that far to overpay for a 5070.
 
My purchasing experience, likely what many others faced:

The launch was an unhappy event for me as a willing buyer. Even though you could get RTX 5070 Ti for MSRP in a 15 minute timeframe from a very popular domestical online store, here in Finland, the store didn't for some reason offer any RTX 5070s - well, one was listed but not orderable. I chose Proshop, a Danish Scandinavian wide store, as my choice of retailer, but hitting the "add to basket" button at 16:00 resulted in an error, unsurprisingly, since there was whole five units of the Prime model in stock! For MSRP you could also order the Inno3D Twin X2, but I don't want such a small cooler and it seems neither do others desire it, since it was available for quite a while. Also the Zotac Solid was orderable for MSRP, but not in stock actually, only arriving at a later date.

Fortunately and somewhat unfortunately I didn't realize immediately that one other store was actually participating in the launch sale, since it seems to me they listed the models not until 16:00, so I likely missed a chance, but they let you leave an order anyway for some of the models, so I'm (likely) getting the TUF variant (non-OC) for the MSRP, which is sort of nice given the price (679 €), but I have to wait for a month for it to arrive, and that's an estimation, and I don't have a dGPU at all. 😭

You should never count on launch availability I guess, nor announced launch dates, since RTX 5070 was already pushed back once. Consider yourself lucky if you wished to purchase and got this SKU, I think it's a decent chip, but should be cheaper of course.
 
Last edited:
Back
Top