Intel faster than AMD, why?

[y1901y]

Why is performance on an AMD processor half as good as on an Intel processor?
After all, AMD processor is more productive by all parameters.

Testing performance on Intel and AMD processors, bench parameters:
Main:
OS - Windows Server 2022
Hypervisor - VMware ESXi 7.0 U3

Stand01:
CPU - AMD EPYC 9374F, 4 vCPU
RAM - 32 Gb

Stand02:
CPU - Intel Xeon Gold 6354, 4 vCPU
RAM - 32 Gb

Results:
 Model | AMD | Intel
medium | 41.46s | 24.92s
large-v2 | 96.58s | 50.47s
large-v3 | 87.21s | 44.89s
turbo | 108.06s | 51.17s

import whisper
import os
import time
from faster_whisper import WhisperModel, BatchedInferencePipeline

test_file = 'audio.mp3'
devices = [ 'cpu' ]#, 'cuda' ]
model_list = [ 'medium', 'large-v2', 'large-v3', 'turbo' ]
compute_types = [ 'int8_float32',  'float32' ] #, 'int8_float16', 'float16', 'bfloat16' ]

print('faster_whisper implementations')
for y in devices:
	for i in model_list:
		for x in compute_types:
			model = WhisperModel(i, device=y, compute_type=x)
			batched_model = BatchedInferencePipeline(model=model)
			
			duration_sum = 0
			start = time.time()
			segments, info = batched_model.transcribe(test_file, language='ru', task='transcribe', batch_size=16)
			end = time.time()
			duration_sum = duration_sum + end - start
			print("{} - quant {} for {} model with fp16 costs {:.2f}s".format(y, x, i, duration_sum))
			del model
#########################
print('whisper implementations')
device = torch.device('cpu')
model_list = [ 'medium', 'large-v2', 'large-v3', 'turbo' ]
fp16_bool = [True, False]
for i in model_list:
	for k in fp16_bool:
		model = whisper.load_model(name=i, device=device)
		duration_sum = 0
		audio = whisper.load_audio('audio.mp3')
		start = time.time()
		result = model.transcribe(audio, language='en', task='transcribe', fp16=k)
		end = time.time()
		duration_sum = duration_sum + end - start
		print("{} model with fp16 {} costs {:.2f}s".format(i, k, duration_sum))
		del model