By using this site, you agree to our Privacy Policy and our Terms of Use. Close
sc94597 on 27 May 2025
Edit Thread

On the conversation of whether or not the Switch 2 Handheld mode can be that power-efficient. 

Here is a table I compiled of Ampere chips. I added different 5W, 6W, and 7W scenario's for the Switch 2 handheld and Switch 2 Docked (15W.) 

First of all, the TSMC 7nm Ampere chips are nowhere near the most power-efficient. Take the A30X for example. It is way down the list, despite being on that node. This illustrates how much the specific GPU/SOC configuration matters and how process node, while mattering, isn't the only factor. 

Secondly, notice how much more efficient the lower TGP chips are. In fact I added a plot showing how lower power consumption for Ampere chips correlates with higher efficiency (excluding the hypothetical Switch 2 efficiencies of course.) 

GPU SKU GPU Core Count Max Clock TFLOPs PerfPerWatt Power Consumption
Switch 2 (Handheld) - 5W TDP T239 1536 561 1.723392 0.3446784 5
GeForce MX570 GA107S 2048 1155 4.73088 0.315392 15
Switch 2 (Handheld) - 6W TDP T239 1536 561 1.723392 0.287232 6
RTX A500 Embedded GA107S 2048 1335 5.46816 0.273408 20
Switch 2 (Handheld) - 7W TDP T239 1536 561 1.723392 0.246198857 7
RTX A5500 Max-Q GA103 7424 1260 18.70848 0.233856 80
A10G GA102 9216 1710 31.51872 0.2101248 150
RTX A500 Mobile GA107S 2048 1537 6.295552 0.209851733 30
GeForce RTX 3080 Ti Max-Q GA103 7424 1125 16.704 0.2088 80
A10 PCIe GA102 9216 1695 31.24224 0.2082816 150
RTX A5000 Max-Q GA104 6144 1350 16.5888 0.20736 80
Switch 2 (Docked) T239 1536 1008 3.0962688 0.20641792 15
GeForce MX570 A GA107SB 2048 1155 4.73088 0.1892352 25
RTX A4500 Embedded GA104 5888 1215 14.30784 0.178848 80
RTX A4500 Max-Q GA104 5888 1215 14.30784 0.178848 80
RTX A4000 Max-Q GA104 5120 1395 14.2848 0.17856 80
RTX A2000 Embedded GA107 2560 1177 6.02624 0.172178286 35
GeForce RTX 2050 Mobile GA107 2048 1245 5.09952 0.169984 30
GeForce RTX 3080 Ti Mobile GA103 7424 1260 18.70848 0.162682435 115
GeForce RTX 3050 Max-Q Refresh 4 GB GA107 2048 1375 5.632 0.160914286 35
GeForce RTX 2050 Max-Q GA107 2048 1155 4.73088 0.157696 30
A10M GA102 7168 1635 23.43936 0.1562624 150
GeForce RTX 3070 Ti Max-Q GA104 5888 1035 12.18816 0.152352 80
RTX A4000 Mobile GA104 5120 1680 17.2032 0.149593043 115
RTX A5500 GA102 10240 1665 34.0992 0.148257391 230
GeForce RTX 3050 Max-Q Refresh 6 GB GA107 2560 990 5.0688 0.144822857 35
GeForce RTX 3070 Ti Mobile GA104 5888 1410 16.60416 0.144384 115
RTX A3000 Mobile GA104 4096 1230 10.07616 0.143945143 70
GeForce RTX 3050 Mobile Refresh 4 GB GA107 2560 1207 6.17984 0.137329778 45
RTX A4000 GA104 6144 1560 19.16928 0.136923429 140
RTX A4000H GA104 6144 1560 19.16928 0.136923429 140
RTX A3000 Mobile 12 GB GA104 4096 1170 9.58464 0.136923429 70
RTX A5500 Mobile GA103 7424 1500 22.272 0.134981818 165
RTX A1000 GA107 2304 1462 6.736896 0.13473792 50
RTX A1000 Embedded GA107S 2048 1140 4.66944 0.133412571 35
RTX A5000 Mobile GA104 6144 1575 19.3536 0.129024 150
RTX A4500 Mobile GA104 5888 1500 17.664 0.126171429 140
GeForce RTX 3050 Max-Q GA107 2048 1057 4.329472 0.1236992 35
GeForce RTX 3050 Mobile GA107 2048 1343 5.500928 0.122242844 45
RTX A500 GA107 2048 1770 7.24992 0.120832 60
RTX A5000 GA102 8192 1695 27.77088 0.120742957 230
RTX A5000-12Q GA102 8192 1695 27.77088 0.120742957 230
RTX A5000-8Q GA102 8192 1695 27.77088 0.120742957 230
RTX A4500 GA102 7168 1650 23.6544 0.118272 200
RTX A2000 GA106 3328 1200 7.9872 0.114102857 70
RTX A2000 12 GB GA106 3328 1200 7.9872 0.114102857 70
GeForce RTX 3050 A Mobile GA106 1792 1343 4.813312 0.106962489 45
RTX A3000 Mobile 12 GB GA104 4096 1440 11.79648 0.102578087 115
GeForce RTX 3050 Mobile Refresh 6 GB GA107 2560 1492 7.63904 0.101853867 75
GeForce RTX 3080 Ti 20 GB GA102 10240 1665 34.0992 0.097426286 350
RTX A1000 Mobile 6 GB GA107 2560 1140 5.8368 0.09728 60
GeForce RTX 3050 6 GB GA107 2304 1470 6.77376 0.096768 70
RTX A2000 Mobile GA107 2560 1687 8.63744 0.090920421 95
GeForce RTX 3090 Ti GA102 10752 1860 39.99744 0.0888832 450
Jetson AGX Orin 64 GB GA10B 2048 1300 5.3248 0.088746667 60
GeForce RTX 3080 12 GB GA102 8960 1710 30.6432 0.087552 350
RTX A2000 Mobile 8 GB GA107 2560 1612 8.25344 0.086878316 95
Jetson Orin Nano 8 GB GA10B 1024 625 1.28 0.085333333 15
Jetson AGX Orin 32 GB GA10B 1792 930 3.33312 0.083328 40
GeForce RTX 3060 Ti GA103 GA103 4864 1665 16.19712 0.0809856 200
GeForce RTX 3080 Ti GA102 10240 1365 27.9552 0.079872 350
GeForce RTX 3050 4 GB GA107 2048 1740 7.12704 0.079189333 90
GeForce RTX 3050 8 GB GA107 GA107 2560 1777 9.09824 0.07911513 115
Jetson Orin NX 8 GB GA10B 1024 765 1.56672 0.078336 20
A800 PCIe 40 GB GA100 6912 1410 19.49184 0.07796736 250
A800 PCIe 80 GB GA100 6912 1410 19.49184 0.07796736 250
RTX A1000 Mobile GA107 2048 1140 4.66944 0.077824 60
A2 GA107 1280 1770 4.5312 0.07552 60
A2 PCIe GA107 1280 1770 4.5312 0.07552 60
GeForce RTX 3070 TiM GA104 5888 1410 16.60416 0.075473455 220
Jetson Orin NX 16 GB GA10B 1024 918 1.880064 0.07520256 25
GeForce RTX 3070 6144SP GA104 6144 1770 21.74976 0.074999172 290
GeForce RTX 3070 Ti GA104 6144 1770 21.74976 0.074999172 290
GeForce RTX 3070 Ti 16 GB GA104 6144 1770 21.74976 0.074999172 290
GeForce RTX 3070 Ti 8 GB GA102 GA102 6144 1770 21.74976 0.074999172 290
GeForce RTX 3060 12 GB GA106 3584 1777 12.737536 0.074926682 170
GeForce RTX 3060 12 GB GA104 GA104 3584 1777 12.737536 0.074926682 170
GeForce RTX 3060 8 GB GA106 3584 1777 12.737536 0.074926682 170
GeForce RTX 3060 Ti GDDR6X GA104 4864 1665 16.19712 0.0719872 225
GeForce RTX 3050 Ti Max-Q GA107 2560 1035 5.2992 0.070656 75
GeForce RTX 3050 Ti Mobile GA106 2560 1035 5.2992 0.070656 75
GeForce RTX 3050 Ti Mobile GA107 2560 1035 5.2992 0.070656 75
GeForce RTX 3050 8 GB GA106 2560 1777 9.09824 0.069986462 130
CMP 90HX GA102 6400 1710 21.888 0.0684 320
A100X GA100 6912 1440 19.90656 0.0663552 300
GeForce RTX 3060 8 GB GA104 GA104 3584 1777 12.737536 0.065320697 195
RTX A2000 Max-Q GA107 2560 1207 6.17984 0.065050947 95
A100 PCIe 80 GB GA100 6912 1410 19.49184 0.0649728 300
Jetson Orin Nano 4 GB GA10B 512 625 0.64 0.064 10
RTX A2000 Max-Q 8 GB GA107 2560 1177 6.02624 0.063434105 95
A30 PCIe GA100 3584 1440 10.32192 0.062557091 165
PG506-207 GA100 3584 1440 10.32192 0.062557091 165
PG506-217 GA100 3584 1440 10.32192 0.062557091 165
PG506-232 GA100 3584 1440 10.32192 0.062557091 165
PG506-242 GA100 3584 1440 10.32192 0.062557091 165
GeForce RTX 3050 OEM GA106 2304 1755 8.08704 0.062208 130
RTX A400 GA107 768 1762 2.706432 0.05412864 50
CMP 70HX GA104 3840 1395 10.7136 0.053568 200
CMP 170HX 10 GB GA100 4480 1410 12.6336 0.0505344 250
CMP 170HX 8 GB GA100 4480 1410 12.6336 0.0505344 250
A800 SXM4 80 GB GA100 6912 1410 19.49184 0.0487296 400
A30X GA100 3584 1440 10.32192 0.044877913 230
A16 PCIe GA107 1280 1755 4.4928 0.0179712 250

Conclusion: Given this correlation between power-consumption and power-efficiency, I don't think the Switch 2 Handheld mode being at the top of the list in efficiency for Ampere is unreasonable, and it isn't too far from other Ampere chips (like the A5500 Max-Q, for example.) 

Last edited by sc94597 - on 27 May 2025