Monday, August 26th 2024

NVIDIA's RTX 5060 "Blackwell" Laptop GPU Comes with 8 GB of GDDR7 Memory Running at 28 Gbps, 25 W Lower TGP

In a recent event hosted by Chinese laptop manufacturer Hasee, company's chairman Wu Haijun unveiled exciting details about NVIDIA's upcoming GeForce RTX 5060 "Blackwell" laptop GPU. Attending the event was industry insider Golden Pig Upgrade, who managed to catch some details of the card set to launch next year. The RTX 5060 is expected to be the first in the market to feature GDDR7 memory, a move that aligns with earlier leaks suggesting NVIDIA's entire Blackwell lineup would adopt this new standard. This upgrade is anticipated to deliver substantial boosts in bandwidth and possibly increased VRAM capacities in other SKUs. Perhaps most intriguing is the reported performance of the RTX 5060. Wu said this laptop SKU could offer performance comparable to the current RTX 4070 laptop GPU. It's said to exceed the RTX 4070 in ray tracing scenarios and match or come close to its rasterization performance.

This leap in capabilities is made even more impressive by the chip's reduced power consumption, with a maximum TGP of 115 W compared to the RTX 4060's 140 W. The reported power efficiency gains are not exclusive to RTX 5060. Wu suggests that the entire Blackwell lineup will see significant reductions in power draw, potentially lowering overall system power consumption by 40 to 50 watts in many Blackwell models. While specific technical details remain limited, it's believed the RTX 5060 will utilize the GB206 GPU die paired with 8 GB of GDDR7 memory, likely running at 28 Gbps in its initial iteration.
Source: via Wccftech
Add your own comment

108 Comments on NVIDIA's RTX 5060 "Blackwell" Laptop GPU Comes with 8 GB of GDDR7 Memory Running at 28 Gbps, 25 W Lower TGP

#26
Chrispy_
DavenThis is an article about mobile meaning laptop 5060 not desktop.
Desktop parts are likely to be 8GB too though - Kopite7kimi leaks are pointing at 128-bit for GB207 and GB206 for xx60 series and Ti variants.

Hell, the 5060 is slated to have 128-bit GDDR6 so it's going to have godawful bandwidth unlike the GDDR7 5060Ti and above which at least might solve the abysmal bandwidth issues the 4060Ti has.
Posted on Reply
#27
Dr. Dro
igormpThey did mention CUDA stuff, if they're running any kind of large-ish ML model, VRAM often is a pretty major limitation compared to raw compute.
Regardless:

6 to 8 GB is still +25% in capacity
8 Gbps to 28 Gbps is +350% in memory bandwidth
Posted on Reply
#28
ttnuagmada
SirB8 GB? Hard pass nvidia. I have 12GB on my 3060. Not downgrading. Pure greed.
There's not a single game you'll play where 8 gigs will be a limiting factor at settings that will be playable.
Posted on Reply
#29
igormp
Dr. DroRegardless:

6 to 8 GB is still +25% in capacity
8 Gbps to 28 Gbps is +350% in memory bandwidth
If your workloads don't fit within 8GB, it's a moot point still.

Anyhow, you're forgetting that the 1060 6gb has a 192-bit bus, that's almost 200GB/s of memory bandwidth with GDDR5. That 5060 with 128-bit should have a total memory bandwidth of ~320GB/s, so a 50~70% increase in bandwidth, quite far from that 350% you got.
Posted on Reply
#30
ARF
Dn't worry, ngreedya will fix the issue by reducing the textures resolution, everything, regardless of the original game quality, will fit in any VRAM they wish. Even 0.8 GB :D


Posted on Reply
#31
Dr. Dro
igormpIf your workloads don't fit within 8GB, it's a moot point still.

Anyhow, you're forgetting that the 1060 6gb has a 192-bit bus, that's almost 200GB/s of memory bandwidth with GDDR5. That 5060 with 128-bit should have a total memory bandwidth of ~320GB/s, so a 50~70% increase in bandwidth, quite far from that 350% you got.
That's bandwidth per pin improvement, not overall bandwidth. However, the 128-bit version of the 3050 can already pull 256 GB/s with ease, so...

It doesn't really matter. The mobile 5060 will utterly outclass any 10 series GPU.
Posted on Reply
#32
Warigator
8 GB of VRAM in 230-370 USD cards since 2016 if you count RX 480 or since 2015 if you count Radeon 390 (which was $329 after it came out)

totally unprecedented situation

imagine 8 MB being the standard for 10 years
Posted on Reply
#33
ArcanisGK507
I haven't bought laptops with dedicated GPUs for a long time, simply because they all have that awful gaming look... I like a more executive design...
Posted on Reply
#34
64K
Warigator8 GB of VRAM in 230-370 USD cards since 2016 if you count RX 480 or since 2015 if you count Radeon 390 (which was $329 after it came out)

totally unprecedented situation

imagine 8 MB being the standard for 10 years
Those weren't entry level laptop GPUs. For example at that time the entry level RX 460 laptop GPU used 4 GB VRAM. Progress is being made. The whole Nvidia VRAM issue generally comes from AMD fans that really only have one thing that they can point to and that's more VRAM than Nvidia. They blow it out of proportion though. Don't buy into the hype.
Posted on Reply
#35
ARF
64KThose weren't entry level laptop GPUs. For example at that time the entry level RX 460 laptop GPU used 4 GB VRAM. Progress is being made. The whole Nvidia VRAM issue generally comes from AMD fans that really only have one thing that they can point to and that's more VRAM than Nvidia. They blow it out of proportion though. Don't buy into the hype.
The community knows better than the nvidia and its shills, that's for sure.

pcmasterrace/comments/1b2wrxa



How much VRAM do I need for Games?
There is no doubt that VRAM is crucial for gaming performance. Not having enough can mean poor performance that includes blurry textures and choppy frames. As games have become more demanding, we are seeing more and more instances where even 8GB is not enough for some modern titles.
Thankfully more than 8GB is becoming more commonplace, and 12GB should be viewed as the recommended minimum amount of VRAM for 1440p or 4K gaming.
www.cyberpowerpc.com/blog/how-much-vram-do-i-have/?srsltid=AfmBOorIfIwlbm5YgSvDo5OZq9x_W0mZ6SDX2OUpbUYL3OeNUcbWR-XH
Posted on Reply
#36
GhostRyder
To be fair I agree with the problems of being stuck at 8gb especially on a new generation card that is considered to be the last line of a mainstream gaming card. But we also have to realize this is more close to a RTX 5050 or 5050 ti desktop card (No idea which since nothing is released yet). So I don't think its unreasonable at least in my book even if I would prefer at least 10gb in this day and age especially on a card that is probably going to be one of the more common "Gaming" laptop GPU's out there.

I mean if nothing else I do think it will hinder its longevity. We have seen higher VRAM cards last longer/improve over time over similar counter parts just because of the extra VRAM headroom.
Posted on Reply
#37
64K
ARFThe community knows better than the nvidia and its shills, that's for sure.

pcmasterrace/comments/1b2wrxa



How much VRAM do I need for Games?
There is no doubt that VRAM is crucial for gaming performance. Not having enough can mean poor performance that includes blurry textures and choppy frames. As games have become more demanding, we are seeing more and more instances where even 8GB is not enough for some modern titles.
Thankfully more than 8GB is becoming more commonplace, and 12GB should be viewed as the recommended minimum amount of VRAM for 1440p or 4K gaming.
www.cyberpowerpc.com/blog/how-much-vram-do-i-have/?srsltid=AfmBOorIfIwlbm5YgSvDo5OZq9x_W0mZ6SDX2OUpbUYL3OeNUcbWR-XH
Using your own link they say 12 GB is recommended minimum for 1440p and 4K. It doesn't say that about entry level 1080p laptop. Also you are using a store to back up your claim that makes money selling you a more expensive card if you will go for it.
Posted on Reply
#38
igormp
Dr. DroThat's bandwidth per pin improvement, not overall bandwidth. However, the 128-bit version of the 3050 can already pull 256 GB/s with ease, so...

It doesn't really matter. The mobile 5060 will utterly outclass any 10 series GPU.
Overall bandwidth is what matters performance-wise.
Anyhow, which version of the 3050 are you talking about? Most of those (specially the mobile ones) are all sub 200GB/s. AFAIK there's no model that does the number you mentioned, and only the 3050 8gb desktop goes past the 200GB/s mark at the same 128bit bus, but still slower than those 256GB/s you mentioned.
Many of the laptops ones are even 96-bit lol

I'm not arguing against the 5060 being way faster, it's just that being faster but not having enough vram to run your task makes it useless at it, no matter how fast it is (reminder that we're not talking about games here).
Posted on Reply
#39
Dr. Dro
igormpOverall bandwidth is what matters performance-wise.
Anyhow, which version of the 3050 are you talking about? Most of those (specially the mobile ones) are all sub 200GB/s. AFAIK there's no model that does the number you mentioned, and only the 3050 8gb desktop goes past the 200GB/s mark at the same 128bit bus, but still slower than those 256GB/s you mentioned.
Many of the laptops ones are even 96-bit lol

I'm not arguing against the 5060 being way faster, it's just that being faster but not having enough vram to run your task makes it useless at it, no matter how fast it is (reminder that we're not talking about games here).
Yeah, I stand corrected. I remembered it being above 200 GB/s, so I guess the 256 number came in mind naturally. This is my 24/7 setting, it's an 80 W variant - 2070 MHz core. It's a Dell G15 5515 with a Ryzen 5600H processor, so runs on PCIe 3.0 (Cezanne limitation). Other two x4 links are used by the NVMe storage. All in all, snappy. No complaints, really.

Posted on Reply
#40
oxrufiioxo
Whats new, if it doesn't have at least a 70 at the end nvidia dgaf they know millions and millions will buy these laptops anyway and just run medium ish settings at 1080p with DLSS lmao.

Amd is even less competitive mindshare wise on laptop than they are on desktop. So basically a one horse race to the bottom.

It's not much better on desktop if you don't spend 500+ gtfo lol.

As long as people keep buying these gimped $3-400 usd cards they'll keep making them. Can't blame Nvidia for rolling in dat $$$.
Posted on Reply
#41
A&P211
SirB8 GB? Hard pass nvidia. I have 12GB on my 3060. Not downgrading. Pure greed.
This is a mobile graphics card, not the desktop version.
Warigator8 GB of VRAM in 230-370 USD cards since 2016 if you count RX 480 or since 2015 if you count Radeon 390 (which was $329 after it came out)

totally unprecedented situation

imagine 8 MB being the standard for 10 years
The rx480 also came in 4gb versions. The memory ran slower at 1750mhz instead of 2000mhz.

The 140w versions in the 4050 to 4070 are misleading, the 4050-4070 only ran at a max of 100-107w.
The 4050, 4060, and 4070 mobile maxed out at 100-107w
Posted on Reply
#42
duckface
8gb? AI need 16gb or more now, a new model flux have 23gb and half version 11~15gb
Posted on Reply
#43
bug
Chrispy_Desktop parts are likely to be 8GB too though - Kopite7kimi leaks are pointing at 128-bit for GB207 and GB206 for xx60 series and Ti variants.

Hell, the 5060 is slated to have 128-bit GDDR6 so it's going to have godawful bandwidth unlike the GDDR7 5060Ti and above which at least might solve the abysmal bandwidth issues the 4060Ti has.
That would be mistake, imho.
4060 was fine with 8GB (we got the 16GB variant, we saw more VRAM does squat for that card). But if this is claimed to offer 4070 level of performance (laptop or not), I'm pretty sure it can handle more than 8GB.

But that's just theory. I don't pay much attention to leaks and put next zero value on them. I'll reserve judgement for when the products land.
Posted on Reply
#44
oxrufiioxo
bugThat would be mistake, imho.
4060 was fine with 8GB (we got the 16GB variant, we saw more VRAM does squat for that card). But if this is claimed to offer 4070 level of performance (laptop or not), I'm pretty sure it can handle more than 8GB.

But that's just theory. I don't pay much attention to leaks and put next zero value on them. I'll reserve judgement for when the products land.
I mean they gave the 4070 laptop 8GB also lol....
Posted on Reply
#45
wolf
Better Than Native
This leap in capabilities is made even more impressive by the chip's reduced power consumption, with a maximum TGP of 115 W compared to the RTX 4060's 140 W.
Anyone else find this odd when the desktop 4060 is 115W? I mean I'm expecting perf/watt gains here just the product stack seems odd.

Always fun to see a swarm of people who wouldn't have bought a 5060 laptop anyway, complain about it :)
Posted on Reply
#46
Onasi
@wolf
So is the laptop version. And the Max-Q variants is 35W. I have no idea where 140 came from. The 4060Ti is 160W. And there is no mobile version.
Posted on Reply
#47
A&P211
wolfAnyone else find this odd when the desktop 4060 is 115W? I mean I'm expecting perf/watt gains here just the product stack seems odd.

Always fun to see a swarm of people who wouldn't have bought a 5060 laptop anyway, complain about it :)
The mobile 4060 never got upto 140w, it was volt limited by Nvidia to 100-107w.
Onasi@wolf
So is the laptop version. And the Max-Q variants is 35W. I have no idea where 140 came from. The 4060Ti is 160W. And there is no mobile version.
The Max-q public relations BS died off years ago.
Posted on Reply
#48
igormp
Onasi@wolf
So is the laptop version. And the Max-Q variants is 35W. I have no idea where 140 came from. The 4060Ti is 160W. And there is no mobile version.
Maybe it's a TDP-up value that OEMs can set up?

But still, by default the mobile config seems to be either 115W or 35W (on the Max-q), as already said above.
Posted on Reply
#49
Onasi
A&P211The Max-q public relations BS died off years ago.
Nah, it’s alive, just in a different form. It’s basically, and forgive me if I am getting this wrong, is now a low power mode that ALL laptop GPUs possess instead of being a separate GPU SKU. And also a shit-ton of other “power saving” tech. Even NV seems to basically have no idea what it is anymore. But it is “AI powered”, so you know it’s good, right?
Posted on Reply
#50
oxrufiioxo
OnasiNah, it’s alive, just in a different form. It’s basically, and forgive me if I am getting this wrong, is now a low power mode that ALL laptop GPUs possess instead of being a separate GPU SKU. And also a shit-ton of other “power saving” tech. Even NV seems to basically have no idea what it is anymore. But it is “AI powered”, so you know it’s good, right?
Yeah they just have TDP ranges now so not only do you have to look at the specific gpu but also the power the laptop allows it to use.... Makes buying the higher end stuff annoying becuase a 4080 laptop can perform drastically different depending on TDP configuration.
Posted on Reply
Add your own comment
Nov 28th, 2024 23:41 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts