kopite7kimi
@kopite7kimi
Followers
26,076
Following
364
Media
38
Statuses
2,925
Explore trending content on Musk Viewer
CDMX
• 162502 Tweets
Clara
• 130972 Tweets
#DebateChilango
• 115338 Tweets
Emma
• 72859 Tweets
Nuggets
• 71523 Tweets
Taboada
• 59618 Tweets
つばさの党
• 50903 Tweets
Murray
• 49126 Tweets
Iztapalapa
• 45747 Tweets
Jokic
• 36510 Tweets
Denver
• 35177 Tweets
Cruz Azul
• 30252 Tweets
Rayados
• 29245 Tweets
家宅捜索
• 27144 Tweets
Pumas
• 26493 Tweets
Tigres
• 25799 Tweets
Monterrey
• 25493 Tweets
#LokSabaElections2024
• 19978 Tweets
Oilers
• 17885 Tweets
Gobert
• 17170 Tweets
バンドリ
• 14969 Tweets
#ElectionDay
• 14736 Tweets
Gambino
• 14625 Tweets
InfoMINGGUAN PakPRABOWO
• 14354 Tweets
Jordi
• 14180 Tweets
Lebih KEREN KitaKOMPAK
• 13871 Tweets
LOSE MY BREATH REMIXES OUT NOW
• 13775 Tweets
Ernestina Godoy
• 12947 Tweets
Pinned Tweet
Miss the Golden Age of Exploration.
Astronomer Vera Rubin at work at the Lowell Observatory in Flagstaff, AZ, 1965. Rubin transformed modern physics and astronomy with her observations showing that galaxies and stars are immersed in the gravitational grip of vast clouds of dark matter.
8
55
380
10
12
100
Continue to update.
RTX 4090, AD102-300-A1, 16384FP32, 384bit 21Gbps 24G GDDR6X, 450W, base 2235 boost 2520 actual max >2750;
RTX 4080, AD103-300-A1, 10240FP32, 256bit 21Gbps 16G GDDR6X, 420W,
RTX 4070, AD104-275-Kx(x is a number)-A1, 7168FP32, 160bit 18Gbps 10G GDDR6, 300W.
90
111
674
OK, let's do a new summary.
RTX 4090, AD102-300, 16128FP32, 21Gbps 24G GDDR6X, 450W, ~2x3090.
I am disappointed with RDNA3.
That's all.
99
90
656
There is an update of RTX 4070.
PG141-SKU331
7680FP32
12G 21Gbps GDDR6X
300W
TSE >11000
64
60
625
The original RTX 4080 12GB will become RTX 4070 Ti instead.
74
51
618
Let's make some updates.
RTX 4080
PG136/139-SKU360
AD103-300-A1
9728FP32
16GB 23Gbps GDDR6X
total card power 340W
55
53
609
RTX 4090 Ti
PG136/139-SKU310
18176FP32
96M L2
24Gbps 24G GDDR6X
total board power ~600W
71
68
573
"the beast"
PG137-SKU0
AD102-450-A1
18176FP32
48G 24Gbps GDDR6X
total board power ~800W
93
75
511
The following is a summary for easy reading. NOT NEWS.
-RTX 4090, PG136/139-SKU330, 16384FP32, 21Gbps 24G, 450W+.
-RTX 4080 16G, PG136/139-SKU360, 9728FP32, 23Gbps 16G, 340W+.
-RTX 4080 12G, PG141-SKU340/341, 7680FP32, 21Gbps 12G, 285W+.
I have finished my work basically.
54
68
491
Some updates.
RTX 4090, AD102-300, 16384FP32, 384bit 21Gbps 24G GDDR6X,
RTX 4080, AD103-300, 10240FP32, 256bit (?18Gbps 16G GDDR6?),
RTX 4070, AD104-275, 7168FP32, 160bit 18Gbps GDDR6 10G.
And DO NOT expect a lower MSRP.
73
70
488
This is a preliminary picture of T234 in Wikipedia. Very clear.
So why do we always guess?
Nintendo will use a customized one, T239.
24
108
473
I'm not a chatterbox, but I have to make some updates. I hope you don't mind.
a possible RTX 4080,
PG136/139-SKU360
AD103-300-A1
9728FP32
256bit 16G 21Gbps GDDR6X
total power ~420W
TSE ~15000
Now I have completed the latest update for 4090, 4080 and 4070.
42
41
422
😂Jensen dropped the MSRP of RTX4090 at the last moment. It was a conservative keynote.
29
20
373
Combined with multiple sources, I confirm the gaming flagship of Ada-next will have a 512-bit memory interface.
48
40
375
RTX 4080, TSE >15000,
RTX 4070, TSE ~10000.
These scores are based on specs I mentioned before.
Honestly, it's not quite sure yet except RTX 4090. Well, I don't care.
36
32
356
RTX 4060 Ti has a very short reference board. The PG190 still uses CEM5 connector.
AD106-350-A1
4352FP32
8G 18Gbps GDDR6
32M L2
220W
51
40
342
A truth. The power limits:
AD102, 800W;
AD103 (DT), 450W, AD103 (Mobile), 175W;
AD104 (DT), 400W, AD104 (Mobile), 175W;
AD106 (DT), 260W, AD106 (Mobile), 140W.
But I don't think we need to use the full power cap.
51
54
325
The problem is power consumption. The power consumption of RTX 4080 and RTX 4070 will be reduced.
28
18
326
Before the keynote, I want to update the spec of RTX 4080 16G.
-22.5Gbps
-start from 320W
After so many months of waiting, these details have become unimportant.
Enjoy.
18
19
319
Ada Lovelace will come out a little bit earlier. Keep patient.
13
23
310
Let's make a summary.
3090S 10752 24GB G6X
3080S 8960 12GB G6X
3070S 5888 8G G6X
3060S 5632 12G G6
Although I doubt the specs of some of them and the name of 90S.
👀👀👀
65
75
308
I don't care about the real release date. I'm just curious about the performance of RTX 4060, which consumes more power than RTX 3070.
37
17
307
The mining perf of 6800 XT is a bit higher than 5700 XT, much lower than RTX 3080.
Is that a good news?
35
15
302
I really don't know their real names.
GA102 20GB GD6X
GA102 12GB GD6X
GA103 10GB GD6X
GA103S 8GB GD6X
GA106 12GB GD6
GA106 6GB GD6
WHAT A MESS!
53
29
302
In fact, there is another full-fat AD102 SKU with 900W TGP, 48G 24Gbps GDDR6X, 2*16pin and higher frequency.
But no one knows whether it will become an actual product. Because the test board of AD102 has more than two 16pin connectors, so everything is possible.
39
42
302
RTX 3080 Ti FE:
PG133-SKU15,
GA102-250-KD-A1,
20GB GD6X,
the same FP32 count as 3090, 10496FP32,
the same MEM speed and TGP as 3080,
no NVLINK.
32
35
284
As I have mentioned before, there is an AD104 SKU with a 400W limit.
PG141-SKU331
a full-fat AD104 with 7680FP32
21Gbps 12G GDDR6X
It can easily match RTX 3090 Ti.
30
22
289
I'm afraid there won't be RTX 4090 Ti anymore. Some low-grade AD103 and AD106 chips will be another versions of RTX 4070 and 4060.
48
42
278
RTX 4060 still uses PG190.
AD107-400-A1
3072FP32
8G GDDR6 18Gbps
115W
24M L2
I will try to remain neutral about any leaks in the future. 😁😁😁
53
23
282
Review the two possible specs of RTX 4070:
PG141-SKU340/341, 7680FP32, 12G 21Gbps GDDR6X, total card power 285W, TSE <11000.
PG141-SKU336/337, 7168FP32, 10G 21Gbps GDDR6X, total card power 250W, TSE <10000.
I found that I had made a typo before, mistaken wrote 341 as 331. 😅
26
22
276
As you can see, only RTX 4090 are relatively certain at present. Be patient. 🙄
24
16
276
As the release approaches, there will be more news about RTX 4090. I just want to remind you once again that AD102 and GA102 have the same pins, especially the design of RTX 3090 Ti can be directly used for RTX 4090. At least no major changes are needed.
Enjoy this process.
19
13
272
Just repeat it again.
As far as I know, the spec of RTX 3080 Ti is 10496FP32, 20G GD6X, the same as I said on November 4th.
Nothing new.
23
24
255
In fact, we got some information about:
The TGP of RTX 4090 is still a mystery. PG139-SKU330 still has the opportunity to achieve 600W.
Possible RTX 4080, PG139-SKU360 has the opportunity of 450W TGP.
If RTX 4070 uses GDDR6X instead, it has opportunity to reach 400W TGP.
23
19
246
A summary at 7.8
Maybe,
RTX 3090 5248 24G GDDR6X 350W
RTX 3080 4352 10G GDDR6X 320W
RTX 3070Ti 3072 8G GDDR6X 250W
RTX 3070 2944 8G GDDR6 220W
For reference only.
22
23
241
I know some people are disappointed with this score. Don't worry, it's just a very conservative score with low TGP, low frequency and L2 cut. 19000+ is just a beginning.
34
5
244
Maybe,
RTX 3090 Super
"GA102-350-A1
10752FP32, no NVLINK
Original 21Gbps MEM
TGP>=450W"
Launch in 2021.
🤫🤫🤫
28
37
242
AMD deserves praise for reaching this level without AI. That's incredible.
Currently testing FSR 2.0. I have to say, wow, great job AMD. 👍
▶️ Deathloop on RTX 3090
▶️ 3840x1600, maxed out settings
4
11
157
10
20
229
If nothing else, we will see 4090, 4080 and 4070 in 2022Q3. I don't think this is new information.
41
24
225
Well, if everything you said OK, we will see:
RTX 4080 Super, AD103-400, 10240FP32;
RTX 4070 Ti Super, AD103-275 or AD102-175, 8448FP32, 48M L2;
RTX 4070 Super, AD104-350 or AD103-175, 7168FP32, 48M L2.
I still doubt with them, especially the Ti Super. I cannot fully agree.
38
45
231
That's why I told you that the rendered picture of TITAN from MLID is totally fake. This is not a simple enlargement of RTX 4090. Please pay attention to its IO part.
NVIDIA’s quad-slot GeForce RTX 4090Ti/TITAN 800W graphics card has been pictured
84
158
688
20
14
227
It's time to rule the world.
Holy shit... 🚨
~15% faster than Comet Lake
~27% faster than Zen 2
23
62
254
8
30
219
Ah, the desktop cards based on GA106 will be called
RTX 3060(3840FP32)
and
RTX 3050 Ti(3584FP32).
😅I don't know if it will change in the future.
23
26
217
Rest in peace, PSU.
I don't want to leak details now.
Let's wait and see.
32
18
208
AD106 is not very strong. The typical TSE score is < 7000.
And both AD106 and AD107 are using PCIE x8.
26
14
215
Some updates of RTX 4070:
PG141-SKU336/337
AD104-250-A1
5888FP32
12G 21Gbps GDDR6X
36M L2
250W
30
29
206
After the dramas of GA100 and GH100, it seems that GB100 is finally going to use MCM.
11
16
214
Things have changed.
Maybe the current spec is 10240FP32, 12G GD6X. 😃
31
21
208
April Fools' Day,
**90 PG139-SKU310, AD102-300, 24G 24Gbps, 600W;
**80 PG139-SKU330, AD102-200, 20G 24Gbps, 500W;
**70 PG141-SKU310, AD104-400, 12G 24Gbps, 400W.
🫣🫣🫣If I am wrong, pls don't kick my ass. 🤣🤣🤣
20
20
201
I was wrong. 🥹
29
8
205
Merry Christmas.
I wish I can confirm the specs of ADA and RDNA3 next year as soon as possible.
10
6
199
Although I don't know your name, I know you there, beast.
AD102-450-A1
15
7
206
Calm down, and focus on the performance of traditional rasterization.
12
15
198
We already know that PCI-SIG will modify the design of 12VHPWR connector.
Jensen will follow up this change soon.
Just keep patient.
18
13
201
RTX 4080 will use AD103 chips, build with 16G GDDR6X, have a similar TGP to GA102.
RTX 4070 will use AD104 chips, build with 12G GDDR6, 300W.
Both of them haven't started testing yet, but soon do them.
29
29
195
I know these before,
AD102 12*6 384bit ~600mm²
NOT MY SOURCES,
AD103 7*6 256bit
AD104 5*6 192bit
AD106 3*6 128bit
AD107 3*4 128bit
Celebrate the leak!
You don't need me anymore. 🤣🤣🤣
11
31
192
Too many annoying political news push, let's talk about something else.
There is no news about RTX 4080 Ti currently, and I don't think you would like to see a 4080 Ti (or Super) based on AD103-400.
Keep patient.
21
12
201
It's quite simple to achieve 2x performance of RTX 3090.
Look forward to further optimization.
25
11
197
waste of sand
10
12
193
Jensen is testing many specs, and the final one has not been determined yet. WAIT.
27
4
191
Wow, more than 10k followers.
It's a great honor to me.
Let's look forward to Hopper with 48V power and 1xxxW consumption.
What a MONSTER!
21
5
185
Although I still have fantasies about 512 bit, the memory interface configuration of GB20x is not much different from that of AD10x.
19
11
188
GB202 will use the same process node as GB100. I must clarify once again that TSMC 4N(vidia) is based on TSMC 5, not 4nm.
I'm sorry I cannot match Jensen's naming with TSMC's naming. We need professional chip analysis to determine.
At least, there is a 30% increase in density.
18
25
192
GA102 has a "7*6" structure.
Maybe AD102 will get a "12*6" structure.
14
18
179
I hope I can hit all the targets, but sometimes the situation is too strange. 🙏
For example, I haven't heard something about RDNA3 since a few months ago, especially the performance update. 🙄
15
5
181
AD102 is still being optimized. It's very simple to reach more than 2xGA102.
So damn simple.
19
7
180
Someone told me the xx90 with 600W TGP has been confirmed.
I think it's too early to talk about it.
27
22
171
@mooreslawisdead
@anthony256
@usmanpirzada
@VideoCardz
@AnilGanti
Totally fake. The Titan doesn't look like this. It has three fans.
17
11
171
I'm sorry I can't provide too much detail about the RDNA2 performance now. Because that is not from my source.
I try to confirm every rumor as soon as possible.
Just wait for leaks. Be patient.
10
2
170
I think the asymmetric design of the 7950X3D is correct.
Maybe it is not the simplest design, but it provides the greatest possibility for gaming.
11
4
172
To be honest, I don't have much information about AMD. Maybe Lisa and Jensen's competition will give us a 100TFLOPS gaming war in a few months.
7
10
170
It is very simple for ADA to reach 3GHz or above. So RTX 4090 has a lower frequency.
11
12
173