Lame, But Cheaper: NVIDIA RTX 4080 Super Review, Benchmark Comparison, & Value Discussion
video description
Date: 2024-01-31
Related videos
Comments and reviews: 20
KYSMO
I highly value the detailed benchmarks and testing provided by Gamers Nexus, and the inclusion of a discount code, STAP, for their store is a welcomed addition. There is optimism that the release of the RTX 4080 Super may lead to a reduction in prices for other GPUs, particularly the RX 7900XTX or XT. The observation is made that international pricing for the card could surpass that of the standard 4080. I suggest incorporating 3D render app performance metrics into GPU reviews. Notably, the 4080 shares the same CUDA cores as the 3080, and the 4080 Super appears to align more closely with the 3080 Ti in terms of core count, potentially explaining the moderate performance gains. Commending the approach of delivering information upfront in videos, I also recommend adopting more visually intuitive comparison charts. I share my decision to choose the Nvidia 4090 over the 4080 Super, considering factors like virtual reality (VR) and concerns about AMD's VR support. Holding Gamers Nexus in high regard for their straightforward approach, I express hope for a reduction in GPU prices. The discussion extends to the possibility of Intel entering the mid-range or high-end GPU market, considerations of the value per frame, and suggestions for improvements in pricing charts. I commend the video quality and seek information regarding export settings. Finally, there is gratitude for Gamers Nexus' commendable work, accompanied by a query about the feasibility of waiting for the 4080 Super in Bulgaria, given potential associated higher costs.
reply
I highly value the detailed benchmarks and testing provided by Gamers Nexus, and the inclusion of a discount code, STAP, for their store is a welcomed addition. There is optimism that the release of the RTX 4080 Super may lead to a reduction in prices for other GPUs, particularly the RX 7900XTX or XT. The observation is made that international pricing for the card could surpass that of the standard 4080. I suggest incorporating 3D render app performance metrics into GPU reviews. Notably, the 4080 shares the same CUDA cores as the 3080, and the 4080 Super appears to align more closely with the 3080 Ti in terms of core count, potentially explaining the moderate performance gains. Commending the approach of delivering information upfront in videos, I also recommend adopting more visually intuitive comparison charts. I share my decision to choose the Nvidia 4090 over the 4080 Super, considering factors like virtual reality (VR) and concerns about AMD's VR support. Holding Gamers Nexus in high regard for their straightforward approach, I express hope for a reduction in GPU prices. The discussion extends to the possibility of Intel entering the mid-range or high-end GPU market, considerations of the value per frame, and suggestions for improvements in pricing charts. I commend the video quality and seek information regarding export settings. Finally, there is gratitude for Gamers Nexus' commendable work, accompanied by a query about the feasibility of waiting for the 4080 Super in Bulgaria, given potential associated higher costs.
reply
fsonetwork
Sadly.. I bought a 4080 at Christmas (The last one I could find..) because I missed out every time a 4090 became available.. so I snatched it up and settled for what I could get. I felt bad about my purchase even then as I'd been trying for months to get a (not outrageously scalped) 4090. Now I just feel like Nvidia sucker punched me in the groin. This is simply getting ridiculous between the years of allowing scalpers (Lets just call them opportunistic criminals from now on. Can we please) to run rampant and holding out on fair MSRP pricing till after most people have given up.
If I didn't utilize some of the extra tools and features Nvidia provides with their cards which AMD nor Intel do not offer I would throw Nvidia the middle finger and switch in a heartbeat. When I make a big upgrade purchase like this I want to feel good about it. I want to be excited to boot up and see my new investment scream out a render or pop with a game I love. Instead I feel nothing but disdain and disappointment as if I've had a gun held to my head and been robbed. My kidnappers have told me to be happy that I'm allowed to pay them for the honor of using their product. Any complaints will always be answered with a shrug and the classic line 'It is what it is'.
reply
Sadly.. I bought a 4080 at Christmas (The last one I could find..) because I missed out every time a 4090 became available.. so I snatched it up and settled for what I could get. I felt bad about my purchase even then as I'd been trying for months to get a (not outrageously scalped) 4090. Now I just feel like Nvidia sucker punched me in the groin. This is simply getting ridiculous between the years of allowing scalpers (Lets just call them opportunistic criminals from now on. Can we please) to run rampant and holding out on fair MSRP pricing till after most people have given up.
If I didn't utilize some of the extra tools and features Nvidia provides with their cards which AMD nor Intel do not offer I would throw Nvidia the middle finger and switch in a heartbeat. When I make a big upgrade purchase like this I want to feel good about it. I want to be excited to boot up and see my new investment scream out a render or pop with a game I love. Instead I feel nothing but disdain and disappointment as if I've had a gun held to my head and been robbed. My kidnappers have told me to be happy that I'm allowed to pay them for the honor of using their product. Any complaints will always be answered with a shrug and the classic line 'It is what it is'.
reply
Kmcornell23
Buy AMD... Buy AMD... Buy AMD... Why Because this is what happens when you don't support price gouging. If we keep supporting AMD we'll eventually get back to a 50/50 market where we as the consumers get the best value we can. If you're considering a 4080, just get the 7900xtx. Most of the time ray tracing is about worthless and the XTX comes with 24gb of VRAM compared to the 16gb that the 4080 has. That's midrange level memory and one of the biggest issues with many Nvidia cards. We vote with our wallets and if you buy Nvidia cards, you're saying it's ok to do the BS that they do. Be part of the solution, not the problem. Then once the market is evened out, buy whatever card you want. But Nvidia needs to be taken down a few notches like what AMD has done to Intel.
reply
Buy AMD... Buy AMD... Buy AMD... Why Because this is what happens when you don't support price gouging. If we keep supporting AMD we'll eventually get back to a 50/50 market where we as the consumers get the best value we can. If you're considering a 4080, just get the 7900xtx. Most of the time ray tracing is about worthless and the XTX comes with 24gb of VRAM compared to the 16gb that the 4080 has. That's midrange level memory and one of the biggest issues with many Nvidia cards. We vote with our wallets and if you buy Nvidia cards, you're saying it's ok to do the BS that they do. Be part of the solution, not the problem. Then once the market is evened out, buy whatever card you want. But Nvidia needs to be taken down a few notches like what AMD has done to Intel.
reply
Frozoken
Using techpowerups numbers i've realised that all the super variants are running at slightly LOWER clock speeds than their base counterparts and the 4070ti super with the biggest regression which makes the performance numbers make much more sense, especially seeing half the consesus for the 4080 super was slightly more cores factory overclock. Using AIB models to normalise clockspeeds via techpowerups reviews (not raw specs), here's the rough performance uplift when the cards are running at the same frequency as their base counterparts: in 4k:
4070 super: 19%
4070ti super: 14%
4080 super: 4%
reply
Using techpowerups numbers i've realised that all the super variants are running at slightly LOWER clock speeds than their base counterparts and the 4070ti super with the biggest regression which makes the performance numbers make much more sense, especially seeing half the consesus for the 4080 super was slightly more cores factory overclock. Using AIB models to normalise clockspeeds via techpowerups reviews (not raw specs), here's the rough performance uplift when the cards are running at the same frequency as their base counterparts: in 4k:
4070 super: 19%
4070ti super: 14%
4080 super: 4%
reply
leod9968
Hey guys, I have to say that this review was way much better than your usual work, which at this point, we can all say it's incredibly valuable. The FPS/$ metric is much useful than, for example, the power efficiency charts or frame latency ones you've recently added to the most recent reviews. I'm not saying those were not interesting but FPS per dollar is a much easier and useful metric we buyers can use to take into account when trying to make up a purchasing decision. I second this new metric, hope you hold onto it and start including it on the reviews that come in the future. Congratulations!
reply
Hey guys, I have to say that this review was way much better than your usual work, which at this point, we can all say it's incredibly valuable. The FPS/$ metric is much useful than, for example, the power efficiency charts or frame latency ones you've recently added to the most recent reviews. I'm not saying those were not interesting but FPS per dollar is a much easier and useful metric we buyers can use to take into account when trying to make up a purchasing decision. I second this new metric, hope you hold onto it and start including it on the reviews that come in the future. Congratulations!
reply
OuijTube
I know you guys want to avoid negative numbers in head to head bar charts. But please, I am begging you, stop. It is better to look at a series of charts and know that bars to the left favor card A and bars to the right favor card B. The way you set your head to bed charts flips which card is on which side of the chart every chart. You might think it makes the individual chart easier to interpret, but it makes interpreting all the charts at once needlessly difficult.
reply
I know you guys want to avoid negative numbers in head to head bar charts. But please, I am begging you, stop. It is better to look at a series of charts and know that bars to the left favor card A and bars to the right favor card B. The way you set your head to bed charts flips which card is on which side of the chart every chart. You might think it makes the individual chart easier to interpret, but it makes interpreting all the charts at once needlessly difficult.
reply
rphoenix5908
I know this one was supposed to be a quick review, but id love to see the 1% lows included. You are already testing the FPS so you should be able to pull the 1% lows from the same test you got the avg FPS from, right So no extra testing needed. Also, dont even have to add a new bar on the chart, if you can just put the 1% low on the same bar (just make it a darker color or something)
reply
I know this one was supposed to be a quick review, but id love to see the 1% lows included. You are already testing the FPS so you should be able to pull the 1% lows from the same test you got the avg FPS from, right So no extra testing needed. Also, dont even have to add a new bar on the chart, if you can just put the 1% low on the same bar (just make it a darker color or something)
reply
gamersnexus
Gn: it's only 1-3% faster for less money
Jtc: they are prettier, but not that much better.
Hub: if not for the silicon, this would literally be nothing but a refresh for a less insane asking price
Smooth brains at Linus tech tips: 800% better! Product of the year! _one-frame insert of Jensen feeding Linus a carrot in the background of the review_
reply
Gn: it's only 1-3% faster for less money
Jtc: they are prettier, but not that much better.
Hub: if not for the silicon, this would literally be nothing but a refresh for a less insane asking price
Smooth brains at Linus tech tips: 800% better! Product of the year! _one-frame insert of Jensen feeding Linus a carrot in the background of the review_
reply
alexfrey4828
Good stuff. Not sure I agree with flipping the charts around the 14:35 mark. As someone who presents data to C-level executives, that would confuse them greatly. Does the audience here pay more attention Better to keep them consistent and just show the how much extra your spending with 4090 on certain games (and relabel from % reduction to % difference).
reply
Good stuff. Not sure I agree with flipping the charts around the 14:35 mark. As someone who presents data to C-level executives, that would confuse them greatly. Does the audience here pay more attention Better to keep them consistent and just show the how much extra your spending with 4090 on certain games (and relabel from % reduction to % difference).
reply
Swarm509
Basically the 4070 TI Super seems like the best choice here if money is a concern and you want to stick with Nvidia. You can can get it cheaper then an 4080, it has 16gb of ram, and one can overclock it to nearly 4080 performance. Might be the sweet spot for Nvidia products at the higher end (and yes everything is expensive, not looking to debate that).
reply
Basically the 4070 TI Super seems like the best choice here if money is a concern and you want to stick with Nvidia. You can can get it cheaper then an 4080, it has 16gb of ram, and one can overclock it to nearly 4080 performance. Might be the sweet spot for Nvidia products at the higher end (and yes everything is expensive, not looking to debate that).
reply
spadac4543
I like that you added the relative cost per fps but prefer how hardware unboxed does $/frame. Gives it a more real feel to the effect of the value of the card. As you mentioned it is still an abstraction of an abstraction and I would shop per price point but it helps show value of say 4060 vs AMD equivalent a little more clearly.
reply
I like that you added the relative cost per fps but prefer how hardware unboxed does $/frame. Gives it a more real feel to the effect of the value of the card. As you mentioned it is still an abstraction of an abstraction and I would shop per price point but it helps show value of say 4060 vs AMD equivalent a little more clearly.
reply
outlet6989
Hi Steve. Have you ever had a GPU card commit suicide while testing it For a future video, I suggest a test of desk mats. It would be informative to review their size, thickness, construction material, and cost. Charts based on these factors might help us when buying one. It would be a good idea not to write comments at 6AM ET.
reply
Hi Steve. Have you ever had a GPU card commit suicide while testing it For a future video, I suggest a test of desk mats. It would be informative to review their size, thickness, construction material, and cost. Charts based on these factors might help us when buying one. It would be a good idea not to write comments at 6AM ET.
reply
noxfelis5333
Here in Sweden, the retail Webhallen has currently the Asus 4070 ti super TUF OC variant gone up to 1209$ from 1080$, the current release price to the Asus 4080 super TUF is at 1300$.
I managed to snagg myself a 4070 ti super OC before the price increase thankfully.
Edit: Here in Sweden we have a 25% VAT on PC parts.
reply
Here in Sweden, the retail Webhallen has currently the Asus 4070 ti super TUF OC variant gone up to 1209$ from 1080$, the current release price to the Asus 4080 super TUF is at 1300$.
I managed to snagg myself a 4070 ti super OC before the price increase thankfully.
Edit: Here in Sweden we have a 25% VAT on PC parts.
reply
boam2943
Thank you for the thirty second review :)
I still watched the rest on the rationale that, if Steve and team wasted their time reviewing it, I should at least watch it. And it could have more surprises in between ;)
Since I only sporadically play minesweeper and solitaire, I think I will wait for the 4050 super launch :P
reply
Thank you for the thirty second review :)
I still watched the rest on the rationale that, if Steve and team wasted their time reviewing it, I should at least watch it. And it could have more surprises in between ;)
Since I only sporadically play minesweeper and solitaire, I think I will wait for the 4050 super launch :P
reply
joesjoeys
As someone who built a new rig with a 4070, I wish Nvidia was more forthcoming earlier that there would be these new refreshes. I wasnt thrilled with the 4070's limitations and I had known that there was a super lineup coming at a more reasonable price point, I likely would have just waited the couple of months.
reply
As someone who built a new rig with a 4070, I wish Nvidia was more forthcoming earlier that there would be these new refreshes. I wasnt thrilled with the 4070's limitations and I had known that there was a super lineup coming at a more reasonable price point, I likely would have just waited the couple of months.
reply
qT_p13
I hope your future reviews dont explain methodology this much. You go on and on sometimes my dude, remember to keep it simple because lots of people don't care/understand half of what you're talking about anyway.
Thanks for the in-depth testing and the time you dedicate to it. It's well appreciated as always.
reply
I hope your future reviews dont explain methodology this much. You go on and on sometimes my dude, remember to keep it simple because lots of people don't care/understand half of what you're talking about anyway.
Thanks for the in-depth testing and the time you dedicate to it. It's well appreciated as always.
reply
Bargate
So basically they have too much pride to just do a normal price cut so they just made a new GPU that is basically the same that is their price reduction. Very sad because in the past super cards were actually pretty decent uplifts of at least 10% could be very misleading to someone not in the know.
reply
So basically they have too much pride to just do a normal price cut so they just made a new GPU that is basically the same that is their price reduction. Very sad because in the past super cards were actually pretty decent uplifts of at least 10% could be very misleading to someone not in the know.
reply
Glotttis
Guys just wait a year or so and pick up 5080 for similar price. Even if it's slightly more expensive, at least you'll get big performance bump, new features, new display ports. There is no reason whatsoever to buy any of these Super cards unless you are building a new PC right now.
reply
Guys just wait a year or so and pick up 5080 for similar price. Even if it's slightly more expensive, at least you'll get big performance bump, new features, new display ports. There is no reason whatsoever to buy any of these Super cards unless you are building a new PC right now.
reply
toddwerther188
You are on 1080p, 1440p, or 4K. Your card choice should align accordingly. IMO, if 1440p either go 60 FPS for RT, or what I suspect most people would do - go 60 FPS rasterization 6650XT or 6750XT and save a boatload of cash for when RT moves beyond ultra enthusiast level.
reply
You are on 1080p, 1440p, or 4K. Your card choice should align accordingly. IMO, if 1440p either go 60 FPS for RT, or what I suspect most people would do - go 60 FPS rasterization 6650XT or 6750XT and save a boatload of cash for when RT moves beyond ultra enthusiast level.
reply
ObakuZenCenter
Yes to 30 minutes of Steve's ever strengthening adjectives over 1.8.....% improvements. You do realize that that could well rise to the dizzying and unprecedented level of a 2.1% improvement with driver refinements. How can anyone not be excited by such a prospect
reply
Yes to 30 minutes of Steve's ever strengthening adjectives over 1.8.....% improvements. You do realize that that could well rise to the dizzying and unprecedented level of a 2.1% improvement with driver refinements. How can anyone not be excited by such a prospect
reply
Add a review, comment
Other channel videos
