Gtx 660m

Likely a dissabled shader unit.

 

Guys the 660M being 128-bit doesn't automatically mean it's slower than the 560M, just because it's a 192-bit card. If bus width is all that matters, the 352 shader, 256-bit GTX 480M would poop all over the 560M, right?

It shouldn't be slower, despite it's position on the chart. The 635M is the rebranded 555M 128-bit, while the 650M is the 28nm mid-high range replacement. Nvidia should now be able to pack the power of the 555M into a 35W card.

The usual. 680M will have more shaders, and maybe more ROPs.

 

ROPs are usually linked into the memory bus.

It would have missing texture units along with the shaders, less tesselation units too.

 

there are two different charts, also the dates refer too the production and the codename was definitely used for the 635m, check the notebookcheckcheck link! also the 610 and 630 are already out with these codenames since they are rebrands! also the 3dmark scores are the same ones as stated before here

my guess is the 650m is the 28nm 540m shrink since its labeled as a perfomance card and not as a enthusiast card such as the 555m or 635m

and why would a 650m be slower than a 635m? because of nvidias crappy naming scheme and marketing

or is the 6550 faster than the 5650?

at best its the 555m shrink or refresh

the charts are no fake since the codenames are eally being used

 

Then why would the 680M which is the same chip as the 675m have a different production date?

 

there are 2 different charts, and the dates maybe fake but on an alienware forum a aw worker claims that the 660m and 675m will be released first and later on the 680m so your actually supporting the dates!

btw the ifo about the m17xr4 using the 660m, 7970m or 675m is also on this forum

 

Now you say they are release dates, but they are not true....

 

Okay first of all I didnt call them release dates, second of all, ive never claimed they were true, I only referred to the codenames

Now im not trying to argue, I was only talking about the 650m and how it shouldve been in the Enthusiast category and not in the perfomance category if its really faster than the 635m which is in the same series

I think its very likely that the codenames are real since the already released 610m, 630m and 635m are being described with them

Im only interested about the smaller cards, since they are the ones that are going to be used in the ultrabooks or sub 14 inch notebooks (soldered on GPUs like the 555m) thats why im speculating about the 650m

In the end this is all speculation, there a little to no facts and I am just trying to use some common sense, to identify what card the 650m could be, I might as well be totally wrong so please dont argue with me

 

I think two things are fairly reliable;

Each part will be sequentially faster than the end-part-number preceding it.

610M -> 630M -> 650M -> 660M -> so on.

And the New Part will be as fast, or faster, than the old Part;

660M is faster than the 560M.

I mean.. that is pretty much a reasonable assumption to make.

 

i think your right!

If the 650m refers to the 550m it could be slower than the 635m, and more like a powersaving oriented card

But I hope the 555m gets an adequate followup, since the 555m is the strongest soldered on card atm, and mxm cards are way too lage for sub 15 inch notebooks

im really optimistic that kepler will be faster and more powersaving, that would be great for mobile gaming

 

the gt 555m the newer one, that was launched in the lg p330 is going to be rebadged as the 635m.

I dont get your point, they are going to launch new cards soon, and since the gt 555m is now the gt 635m, it at least leads us to believe that the 650m is going to be faster. and necessarily a new card.

 

yes but isnt it weird that the 650m is labeled as a perfomance class card while the 635m is labeled as a Enthusiast card?

And following your logic, even the 640m is faster than the 555m, which it probably isnt since the 640m is the followup of the 540m card

Now even if codenames are wrong, im still wondering if the 650m is a soldered-on card on the mainboard, because if it is, this would be released in smaller notebooks than 15 inches

Like I said I am just speculating, is that wrong? ^^

 

nope its just that the logic aint sound

 

it isnt my logic, if you have been following nvidias naming scheme then you come to the conclusion that it doesnt make any sense ^^

btw check this out



doesnt seem so strong for the 650m (n13p-gt)

This was seen at the ces on a new samsung preview notebook

http://www.brightsideofnews.com/news/2012/1/27/samsung-laptops-show-new-nvidia-kepler-mobile-gpus.aspx

 

I'd like the GTX 660M (45W TDP) with a i7-3712QM (BGA, 35W TDP) in a 17.3" laptop chassis.

This would make for a total of 80W TDP laptop that is faster (15% CPU/NVIDIA GPU and 50% Intel IGP) than a previous gen i7-2630QM (45W TDP) + GTX 560M (75W TDP) = 120W TDP total.

 

I think where the logic issue comes into play is the that everyone expects for the 650M to be Faster than the 635M. (Which i think it will be, logically)

Although nvidia may list a particular part "Performance" or "Enthusiast", those terms are simply market-hype and nothing more, and will not be indicative of the spec of the product.

 

yes but check out the picture, the 555m has 144 cuda cores and the 550m has 96, so it could really be that the 650m is the followup of the 550m and not 555m

ofcourse we dont know how the new architecture works maybe its still faster

 

We really have no idea what card that is, and there's no guarantee that the codenames we have are even accurate, as a significant amount of time has passed since any real news about mobile Kepler has surfaced.

Do you think the specs in that picture make sense for a card titled GT 650M? I don't.

 

^^ this. Also nvidia stated that they expected much more from amd 7970, clearly tells us kepler will be kinda good, still I don't think +30% better, but I am now convinced it can be 20-25% beating previous generation

 

I think it's the case of last generation's enthusiast GPU being a current generation performance GPU.
I'm one to believe that GT 650M > GT 635M. Not because of codenames, leaks or anything like that. Just pure common sense. It'd be really confusing for the general public otherwise (imagine a five dollar bill being more money than a ten..). Of course, nothing is written in stone.
What kind of performance can we expect from the GT 650M? Following the "common sense" theory, something between GT 555M and GTX 560M. Closer to which? I don't know. Hope that it's the latter but it's probably not.

Cheers!

 

Actually I think the GT 635M should be faster or very close to the GT 650M.
635M is no Kepler but a rebrand and is basicly a rebranded 555M and with two versions, one with 128bit and another one with 192 bit memory bus

The GT 555M is like 30% (sometimes more) faster than 550M. I don`t know how much faster the 650M will be compared to 550M but it have to perform quite a bit to surpass the 635M (555M).

 

If the 635M ends up being faster than the 640M and 650M, Nvidia will have crossed into a new plane of ridiculousness.

You want to talk about confusing customers? Bigger means better, not slower.

 

Fully agreed

 

If you compare the picture of the N13P-GT (650m kepler codename) with the specs of the N12P-GT (550M), you might see some similiarities... Also this underlines that the codename n13p-gt might be very real!

Check out the text from the source of the picture!

Source: Samsung Laptops Show New NVIDIA 600 Series Mobile GPUs - Bright Side Of News*

In the original article they compare the specs to a 630m, which is in another samsung notebook next to the n13p-gt notebook.. But now compare the specs to the N12P-GT

The 550m is basically an overclocked 540m (which the 630m is a rebrand of) thats why the similiarities in the article

Code:

GeForce GT 550M 96@740MHz
Codename	N12P-GT
Pipelines	96 - unified
Core Speed *	740 MHz
Shader Speed *	1480 MHz
Memory Speed *	900 MHz
Memory Bus Width	128 Bit
Memory Type	DDR3/GDDR5
Max. Amount of Memory	1536 MB
Shared Memory	no
DirectX	DirectX 11, Shader 5.0
technology	40 nm
Notebook Size	large
Date of Announcement	06.01.2011

So if we look at the charts again, we see that the refresh or followup of the 555m (codename N12E-GE-B) isnt actually the 650m, but rather the 660m (Codename N13E-GE)!


[/QUOTE]

Only true if the codenames are real, but the article from the samsung notebooks may indicate that they could certainly be true

 

I see what you are saying.

The 555M is now the 635M.

I would figure the 650M will be the replacement for both the 550M AND 555M from the last generation.. And i think a lot of us figure it will be on-par with the 560M from the last generation.

The 660M/670M/675M and Later 680M will improve upon the segments that the 500 series GPUs all occupied.

 

oh yes, craziness, I hope they will keep up the bigger the better notion for the sake of all

 

guys, im not hoping that the 650m will be weaker then the 635, and obviously the 640m and 650m should be stronger since they have bigger numbers

but for me it looks like the 650m is just a die shrink of the 550m, I do think it brings more perfomance, but how much more is the question? It looks like kepler is faster clock-per-clock but atm we can only guess how much faster kepler really is.. We will just have to wait and see

And like I said, based on the facts I have posted, I think the real follow up for the 555m isnt the 650m, but rather the 660m, since the Codename and TDP fit the description.

BTW, another problem might be availability

Check out these current article on NVIDIA

http://semiaccurate.com/2012/02/15/nvidia-comes-somewhat-clean-on-their-quarterly-conference-call/

 

I don't think anyone thought you were advocating for a backward naming-convention. I think we were just confused a bit.

nVidia has a very long, proud, tradition of screwing the pooch on product names. I mean.. Surely an 8800-GTX is Faster than a GTX-280? Right? oh.. Wait.. How about a GeForce 405, has to be a LOT faster than a GeForce 295.. I mean.. its a whopping 110 more!!

HA!

 

As far as bigger=better goes, you have to understand one thing, which is that the first digit can more or less be ignored when comparing unless you're looking at cards from a good few years back. For example, a 580 is only a little faster than a 480, but the 480 would completely destroy a 520.

 

via Clevo:



Believable? Discuss.

 

LOL I`m gonna eat my own words if that is true. Sick performance gain from Fermi

If it is indeed true then 635M is now 555M, 650M is now somewhere between 560M and 570M since 560M scores 9k and 570M scores 11k. God knows where 660M, 670M and 680M will be. Well I`m guessing that the 660M will be very close to the 580M

And you Kevin: Believable?

 

Completely believable if Nvidia did not screw up the naming. Between GT 635m (or the GT 555m) there is a GT 640M, so performance is just about right, especially when it uses GDDR5. However, you know idiosyncratic manufacturers will probably pair 650M with DDR3 like in the Samsung Series 5 NP550P7 Notebook.

 

Sure, but I'd like to see it confirmed that they were tested with the same CPU.

IF the gain is real, it's won't be shocking, to me. I'm the one who's been predicting huge performance gains from every level of the 28nm transition.

All of these "10 to 20 percent" max people will see the truth, eventually.

 

well if its true that would be great

Then there can be notebooks which are really worthy of being called portable

I soo hope this is true, I would be delighted !

Now please just bring out the lg p330 with ivy bridge and kepler !

This gain would be sick though, its as fast as a 570m?

Really astonishing, just think about the desktop cards, they have to be extremely fast then

BTW what is the official source for that? Is it officially from clevo?

EDIT: Btw compare the specs from the samsung n13p-gt with the 3dmark score, wouldn that be an unbelievable perf/watt ratio?

 

This new information is telling me that the card in the graph and the one in the Samsung are not the same, or the Samsung has a severely crippled version.

Graph came from this post. He always has the first info out of Taiwan. New 11.6" Clevo appears to have been the test platform.

Again, there are still questions to be answered before we get too excited..

 

It wouldn't surprise me if Nvidia pulled the 'crippled version' stunt - they have a tendency to do that (I seriously don't see the point in it, aside of course from mucking the general populace up for the most part and profiteering from the confusion), but we should wait and see before making any conclusions.

70% performance gain is a serious increase indeed - that would be a great edition for the entry 'high end' graphics in the mobile segment - while the remaining two would of course be more powerful.

 

It is not 70% performance gain, it is atleast 100%. If you look at the scores @ notebookcheck you can see a laptop with 2630QM and 550M scoring 4800 in Vantage. 650M is over 10 000 in Vantage, and I don`t think the CPU they have been testing with is any stronger than that. Especially when it is inside a 11" laptop.
Link to notebookcheck http://www.notebookcheck.net/NVIDIA-GeForce-GT-550M.42264.0.html

That is what I have trouble believing. 2x the performance. Sick if true

 

well it is unbelievable !!

How much longer should we wait? this is really getting interesting now

 

People at the forum where this score originated from are questioning the vantage score of the 650M btw. Atleast that what I get from the crappy translation
The GPU have DDR3 memory too, and not GDDR5

Google Oversetter

 

the 10 000 vantage score is also from this picture

 

So the Vantage score is just the usual Nvidia provided marketing propaganda, with hard to depict graph and everything. That's why I said be hesitant.

It quite possibly could be significantly more powerful than the 555M, but let's wait until we have...

1. - the specs of the card. I need a visual GPU-Z reading. Also, is there a GDDR5 version? This needs to be sorted as well.
2. - hand run 3DMark tests, with screenshots and everything

 

btw here are some more pictures of the 11.6 inch notebook and the insides

 

Yes I agree that we should wait to find out more. From the source I gave it is only the DDR3 version. However the ancient pic gamba posted shows that there is also a GDDR5 version, so I don`t really know.
There is no GPU-Z snapshot available yet, perhaps it will come in that thread later

 

so they are about doubling the performance with the previous generation, while amd is about 20-30% (in desktop environment btw), I guess TSMC is selling super-28nm components to nvidia while giving the joke to amd, highly doubt it, otherwise I don't think amd can screw up something so bad while nvidia excels in, they have pretty much the same tech anyway..

I DO hope that we will have about %50 gain in performance, I am buying gtx 680m anyway (already saved the money for it ), but I doubt it because even if they had such a revolutionary product at hand, a profit oriented company won't release it all of a sudden, they will make money from somewhere in between performance (my case for example, I will buy it anyway), why waste such an achievement so quickly?

 

Going by NVIDIA's current situation, if they don't release a revolutionary product in 3 months time, they are in very deep trouble. Namely Intel attacking entry Quadro, and Tegra being part of a tablet fad.

 

Any chance on getting a release date with the 670m or 675m?

I will hold off on my new notebook if better technology is >90 days away.

 

Some say April, concurrent release with Ivy Bridge. How true that is is unverified.

 

You know that, I know that.. but a lot of average users that just want to play Rift or casually game have no clue.

 

Well now.. TSMC is just churning out what AMD and NVIDIA give them. So there is no doubt that each manufacturer could have radically different performance levels of products on the same process. Its happened before.

 

I could see it easily having that much more performance.

You have 28nm vs 40nm which allows for ~25% more cores without heat/power consumption increase. Then with architectural changes you could make up the rest of the increase.

In the meantime I will have to make do on an increasingly ageing 5650. The next however-many-years I am forced to use it before being able to afford a new laptop are going to be a struggle. But as long as I can play Half Life 3 in 1080p at the same visual detail level of HLEP2 then I will be happy.