Huge AMD Navi leak sets Internet alight in run-up to announcement

It’s been a difficult year for hardware manufacturers for sure but AMD have ridden the wave and continue to go from strength to strength and now if the tech specs that have been leaked for the upcoming Navi 21, 22, and 23 GPUs turn out to be accurate, AMD will be steaming into 2021 in an impressive position.

Reddit user STBLR posted the leak after what seems to be some cheeky reverse-engineering from drivers present in the macOS code.

So far AMD has only confirmed they will be announcing the Radeon RX6000 series although they are allegedly developing three new Navi 2X GPUS – the Navi 21, 22, and 23. Big Navi will be Navi 21 and it is expected that the higher numbers will be the lower-end GPUs in the lineup.

The Navi 21 – likely the dual-slot RX6090XT flagship card, looks from the leaker’s table, reproduced below,  that it will have 22.5TFLOPS of compute power based on the leaked value of 80CU, which coupled with 16GB of vRAM will enable it to rip through any game thrown at it. The variable may well turn out to be the price. Has AMD managed to get this affordable?

Uzman Pirzada of tech site WCCFTech said: “Assuming the ratio of CU to SP remains the same we are looking at 5120 stream processors. Now Navi 21 appears to have two further variants, 21a and 21b with the former clocking up to 2.2 GHz. This would put the single-precision compute performance of the cards at a solid 22.5 TFLOPs. For comparison, the NVIDIA RTX 3080 has a single-precision compute performance of 29.8 TFLOPs. This card, if priced correctly by AMD ,could be an absolute winner. ”

We will find out at the end of October.

 

Property	Navi 10	Navi 14	Navi 12	Navi 21	Navi 22	Navi 23	Navi 31
num_se	2	1	2	4	2	2	4
num_cu_per_sh	10	12	10	10	10	8	10
num_sh_per_se	2	2	2	2	2	2	2
num_rb_per_se	8	8	8	4	4	4	4
num_tccs	16	8	16	16	12	8	16
num_gprs	1024	1024	1024	1024	1024	1024	1024
num_max_gs_thds	32	32	32	32	32	32	32
gs_table_depth	32	32	32	32	32	32	32
gsprim_buff_depth	1792	1792	1792	1792	1792	1792	1792
parameter_cache_depth	1024	1024	1024	1024	1024	1024	1024
double_offchip_lds_buffer	1	1	1	1	1	1	1
wave_size	32	32	32	32	32	32	32
max_waves_per_simd	20	20	20	16	16	16	16
max_scratch_slots_per_cu	32	32	32	32	32	32	32
lds_size	64	64	64	64	64	64	64
num_sc_per_sh	1	1	1	1	1	1	1
num_packer_per_sc	2	2	2	4	4	4	4
num_gl2a	N/A	N/A	N/A	4	2	2	4
unknown0	N/A	N/A	N/A	10	10	8	10
unknown1	N/A	N/A	N/A	16	12	8	16
unknown2	N/A	N/A	N/A	80	40	32	80
num_cus (computed)	40	24	40	80	40	32	80