Nvidia 4000 series cards

[enthusiast29]

Its nice of you to think in that way but the answer is NO.

Even a RTX 3060Ti cannot compar to RTX 4080 12 GB. Check the Cuda cores - almost double.

If you knew anything about Nvidia cards history it's that,
1) You cannot compare CUDA core count across different architectures. It's just not the same thing and not in 1:1 comparison ratio.
2) The GPU chip codename is AD104 which means it's a XX60ti/XX70 based GPU as per history of the lineup.

1070 = GP104
2070S = TU104
3060Ti = GA104
3070 = GA104

Here's the data to prove it,

NVIDIA GeForce GTX 1070 Specs

NVIDIA GP104, 1683 MHz, 1920 Cores, 120 TMUs, 64 ROPs, 8192 MB GDDR5, 2002 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 2070 SUPER Specs

NVIDIA TU104, 1770 MHz, 2560 Cores, 160 TMUs, 64 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3070 Specs

NVIDIA GA104, 1725 MHz, 5888 Cores, 184 TMUs, 96 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 4080 12 GB Specs

NVIDIA AD104, 2610 MHz, 7680 Cores, 240 TMUs, 80 ROPs, 12288 MB GDDR6X, 1313 MHz, 192 bit

www.techpowerup.com

Another thing to note, the "Ti" variants of the card are NOT better binned SKUs of the non-Ti variant. It's actually a bad binned SKU of the higher SKU which means that the 3060Ti was meant to be a 3070 but for some reason the yield was not good enough so they called it a 3060Ti and put it up for sale. That's how you not waste Chip wafers. How do I know that? Same thing, look at the chip code name, both called GA104 and the 3060 is actually a whole another chip called GA106.

 

[Fenix]

If you knew anything about Nvidia cards history it's that,
1) You cannot compare CUDA core count across different architectures. It's just not the same thing and not in 1:1 comparison ratio.
2) The GPU chip codename is AD104 which means it's a XX60ti/XX70 based GPU as per history of the lineup.

1070 = GP104
2070S = TU104
3060Ti = GA104
3070 = GA104

Here's the data to prove it,

NVIDIA GeForce GTX 1070 Specs

NVIDIA GP104, 1683 MHz, 1920 Cores, 120 TMUs, 64 ROPs, 8192 MB GDDR5, 2002 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 2070 SUPER Specs

NVIDIA TU104, 1770 MHz, 2560 Cores, 160 TMUs, 64 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3070 Specs

NVIDIA GA104, 1725 MHz, 5888 Cores, 184 TMUs, 96 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 4080 12 GB Specs

NVIDIA AD104, 2610 MHz, 7680 Cores, 240 TMUs, 80 ROPs, 12288 MB GDDR6X, 1313 MHz, 192 bit

www.techpowerup.com

Another thing to note, the "Ti" variants of the card are NOT better binned SKUs of the non-Ti variant. It's actually a bad binned SKU of the higher SKU which means that the 3060Ti was meant to be a 3070 but for some reason the yield was not good enough so they called it a 3060Ti and put it up for sale. That's how you not waste Chip wafers. How do I know that? Same thing, look at the chip code name, both called GA104 and the 3060 is actually a whole another chip called GA106.

So what you are implying is that RTX 4080 12 GB is the successor of RTX 3060, someone did the comparison on the basis of 12GB VRAM. And CUDA cannot be ignored, they matter. Ti are better binned cards not because I have a Ti. 3070 is a circumsized 3070Ti. The 3070Ti is the actual 3070 and 3070Ti could have been binned better but then Green Team did not see any competition. 3080Ti is a better binned 3080 or you can say that 3080Ti should have been the real 3080 and so is the case for 3090Ti. Then AMD followed the league with 6750XT/6650XT and 6950XT.

If I dont see a competition and I am selling the product well, why would I want to change. That is the whole point.

 

[buBleZ]

So any idea when is 4090FE going to be available on RPtech site? I have filled that form they have on its page and waiting.

Don't know why. Nvidia india page for RTX 4000 series points to pcstudio.in - Ankit Infotech, Bangalore. May be for partner cards

 

[enthusiast29]

So what you are implying is that RTX 4080 12 GB is the successor of RTX 3060, someone did the comparison on the basis of 12GB VRAM. And CUDA cannot be ignored, they matter. Ti are better binned cards not because I have a Ti. 3070 is a circumsized 3070Ti. The 3070Ti is the actual 3070 and 3070Ti could have been binned better but then Green Team did not see any competition. 3080Ti is a better binned 3080 or you can say that 3080Ti should have been the real 3080 and so is the case for 3090Ti. Then AMD followed the league with 6750XT/6650XT and 6950XT.

If I dont see a competition and I am selling the product well, why would I want to change. That is the whole point.

That's the way companies want the end user to see it and you fell right into the trap. Anyways I'm not here to argue.
I'm not saying 4080 is a successor to 3060, you'd have known that if you read my post carefully and followed the links but I guess you don't have the time so I know I'm a fool trying to prove something to someone who may not wanna listen at all.
4080 12GB is in-fact a successor to 3060ti/3070 card as per the chip codename, period. They wanted to inflate the price of a XX70 class card due to the production costs (+ maybe greed) and the only way they could fool the consumer is to call it a XX80 class card so they can justify the $300 price hike.

Peace!

 

[buBleZ]

If you knew anything about Nvidia cards history it's that,
1) You cannot compare CUDA core count across different architectures. It's just not the same thing and not in 1:1 comparison ratio.
2) The GPU chip codename is AD104 which means it's a XX60ti/XX70 based GPU as per history of the lineup.

1070 = GP104
2070S = TU104
3060Ti = GA104
3070 = GA104

Here's the data to prove it,

NVIDIA GeForce GTX 1070 Specs

NVIDIA GP104, 1683 MHz, 1920 Cores, 120 TMUs, 64 ROPs, 8192 MB GDDR5, 2002 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 2070 SUPER Specs

NVIDIA TU104, 1770 MHz, 2560 Cores, 160 TMUs, 64 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3070 Specs

NVIDIA GA104, 1725 MHz, 5888 Cores, 184 TMUs, 96 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 4080 12 GB Specs

NVIDIA AD104, 2610 MHz, 7680 Cores, 240 TMUs, 80 ROPs, 12288 MB GDDR6X, 1313 MHz, 192 bit

www.techpowerup.com

Another thing to note, the "Ti" variants of the card are NOT better binned SKUs of the non-Ti variant. It's actually a bad binned SKU of the higher SKU which means that the 3060Ti was meant to be a 3070 but for some reason the yield was not good enough so they called it a 3060Ti and put it up for sale. That's how you not waste Chip wafers. How do I know that? Same thing, look at the chip code name, both called GA104 and the 3060 is actually a whole another chip called GA106.

As a chip design engineer agree with the strategy explained here. All semiconductor companies have a binning strategy and this is Nvidia's

 

[Fenix]

That's the way companies want the end user to see it and you fell right into the trap. Anyways I'm not here to argue.
I'm not saying 4080 is a successor to 3060, you'd have known that if you read my post carefully and followed the links but I guess you don't have the time so I know I'm a fool trying to prove something to someone who may not wanna listen at all.
4080 12GB is in-fact a successor to 3060ti/3070 card as per the chip codename, period. They wanted to inflate the price of a XX70 class card due to the production costs (+ maybe greed) and the only way they could fool the consumer is to call it a XX80 class card so they can justify the $300 price hike.

Peace!

No competition from AMD is the reason.

NVIDIA Publishes Official GeForce RTX 4080 16 GB & RTX 4080 12 GB Gaming Benchmarks, Huge Performance Gaps In Entire RTX 40 Lineup

NVIDIA has published the official gaming benchmarks of its GeForce RTX 4080 16 GB & 12 GB models showing a huge performance gap.

wccftech.com

Take this with a grain of salt.

 

[r4hul]

If ti version comes with pcie 5.0 and dp 2.0 i might go for that. For now waiting for AMD launch as they are having dp 2.0 connection.

 

[Fenix]

If ti version comes with pcie 5.0 and dp 2.0 i might go for that. For now waiting for AMD launch as they are having dp 2.0 connection.

It will take time, we still don't have vfm hdmi 2.1 monitors yet.

 

[r4hul]

Yes I know but when arc can come with dp 2.0 and mostly amd rdna3 will also come so why not a 1.5l card. This was a shitty move by nvidia. Just like where they said 10gb vram is good but then looking at competition now giving more vram.
If I am spending that much then I want it to have all the latest and greatest present even though i wint be able to use it for some time.

 

[enthusiast29]

If I am spending that much then I want it to have all the latest and greatest present even though i wint be able to use it for some time.

By Nvidia's mindset if someone is spending that much to get the latest and greatest they'll spend again next year and that's how you leech the consumer who doesn't think twice.

 

[r4hul]

By Nvidia's mindset if someone is spending that much to get the latest and greatest they'll spend again next year and that's how you leech the consumer who doesn't think twice.

And the customers who have blind money. Damn so many people bought the 4090 already. Where as I'm thinking to change pads of my older card.

 

[PradUp]

RTX 4080 12GB Cancelled before launch.

https://videocardz.com/newz/nvidia-cancels-geforce-rtx-4080-12gb

Unlaunching The 12GB 4080

16GB GeForce RTX 4080 on track to delight gamers everywhere November 16th.

www.nvidia.com

 

[enthusiast29]

RTX 4080 12GB Cancelled before launch.

https://videocardz.com/newz/nvidia-cancels-geforce-rtx-4080-12gb

Awesome!

 

[dexterz]

RTX 4080 12GB Cancelled before launch.

https://videocardz.com/newz/nvidia-cancels-geforce-rtx-4080-12gb

Guess they have to print new labels to stick on these units

 

[enthusiast29]

Guess they have to print new labels to stick on these units

And re-flash with an updated firmware. Imagine the loss to the AIBs who may have got this surprise now.

 

[john1911]

4080 is even printed/etched on the GPU body itself.

 

[Fenix]

RTX 4080 12GB Cancelled before launch.

https://videocardz.com/newz/nvidia-cancels-geforce-rtx-4080-12gb

RTX 4080 12 GB is based on the smaller "AD104" silicon, has just 7,680 CUDA cores (21% fewer CUDA cores); and a meager 192-bit wide GDDR6X memory bus. This had the potential to confuse buyers, especially given the $900 price.

With criticism spanning not just social media but also bad press, NVIDIA decided to pull the plug on the RTX 4080 12 GB. The company will likely re-brand it as a successor to the RTX 3070 Ti, although then it will have a hard time justifying its $900 price-tag. The RTX 4080 16 GB, however, is on track for a November 16 availability date, with a baseline price of $1,200.

It will be relaunched or tweaked.

 

[enthusiast29]

RTX 4080 12 GB is based on the smaller "AD104" silicon, has just 7,680 CUDA cores (21% fewer CUDA cores); and a meager 192-bit wide GDDR6X memory bus. This had the potential to confuse buyers, especially given the $900 price.

With criticism spanning not just social media but also bad press, NVIDIA decided to pull the plug on the RTX 4080 12 GB. The company will likely re-brand it as a successor to the RTX 3070 Ti, although then it will have a hard time justifying its $900 price-tag. The RTX 4080 16 GB, however, is on track for a November 16 availability date, with a baseline price of $1,200.

It will be relaunched or tweaked.

This is exactly what I was trying to explain you that the AD104 is a XX70/60Ti class GPU and the other guy you replied to who said it was really a 4060 was also partially right. This card was only renamed to "4080" to justify the $300 price hike and to fool the consumers but perhaps they forgot about the media.
Now we all know those chips won't be wasted but renamed and resold and definitely for a lower price tag since anything below a 4080 won't be priced the same as "4080 12GB" or the backlash will be more.

BTW the 4080 16GB is also not priced fairly. If it's really a XX80 class GPU then previous MSRP was $700 and the new is $1200 which is more than 70% price hike! So not worth it.

 

[burntwingzZz]

the new reprint card would be the goto card an there would be multiple hacks majorly some AIB might leak internal firmware BIOS if nvidia nerfs it they .Because laser cutting the chip ship has sailed .They should have actually lowered their price and market that 4080 now more getable and so on . have X050 X X060 X070 X080 cutdown version X080 SE than ti 4090 and a titan

 

[Psycharge]

RTX 4080 12GB Cancelled before launch.

Did you mean "Unlaunched", yeah real smooth Nvidia, real smooth.
I don't think this is due to backlash in price to performance because no is saying that the 4080 16GB is priced unfairly but instead due to people realising that the GPU die class is of 70 series.
Also real proffesional of Nvidia to call it unlaunched and delay it's release into November and we know for sure that if AMD doesn't offer tangible competition, Nvidia will just rebrand this as the 4070 and sell it for exactly the same amount.