https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&feed=atom&action=history
nvidia/tegra/xavier - Revision history
2024-03-29T06:11:32Z
Revision history for this page on the wiki
MediaWiki 1.28.1
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=94785&oldid=prev
199.116.118.167: /* GPU */ Fixed tensor core MACs per cycle numbers.
2019-12-09T07:08:16Z
<p><span dir="auto"><span class="autocomment">GPU: </span> Fixed tensor core MACs per cycle numbers.</span></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 07:08, 9 December 2019</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, particularly improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that each Volta multiprocessor contains eight tensor cores each of which can perform <del class="diffchange diffchange-inline">16x </del>FP16 MACs or <del class="diffchange diffchange-inline">32x </del>INT8 MACs per cycle. All of this yields a maximum 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, particularly improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that each Volta multiprocessor contains eight tensor cores<ins class="diffchange diffchange-inline">, </ins>each of which can perform <ins class="diffchange diffchange-inline">64x </ins>FP16 MACs or <ins class="diffchange diffchange-inline">128x </ins>INT8 MACs per cycle. All of this yields a maximum 22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
199.116.118.167
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=89282&oldid=prev
David at 01:50, 27 April 2019
2019-04-27T01:50:52Z
<p></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 01:50, 27 April 2019</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l190" >Line 190:</td>
<td colspan="2" class="diff-lineno">Line 190:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>== Documents ==</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>== Documents ==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* [[:File:ces2018 - nvidia drive xavier.pdf|CES 2018: Nvidia Drive Xavier]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* [[:File:ces2018 - nvidia drive xavier.pdf|CES 2018: Nvidia Drive Xavier]]</div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">== See also ==</ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">* Tesla {{teslacar|FSD Chip}}</ins></div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>== Bibliography ==</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>== Bibliography ==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* IEEE Hot Chips 30 Symposium (HCS) 2018.</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* IEEE Hot Chips 30 Symposium (HCS) 2018.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* Schor, David. (September, 2018). "[https://fuse.wikichip.org/news/1618/hot-chips-30-nvidia-xavier-soc/ Hot Chips 30: Nvidia Xavier SoC]"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>* Schor, David. (September, 2018). "[https://fuse.wikichip.org/news/1618/hot-chips-30-nvidia-xavier-soc/ Hot Chips 30: Nvidia Xavier SoC]"</div></td></tr>
</table>
David
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=85022&oldid=prev
141.113.3.11: /* Deep Learning Accelerator */
2018-12-17T15:23:19Z
<p><span dir="auto"><span class="autocomment">Deep Learning Accelerator</span></span></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 15:23, 17 December 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l96" >Line 96:</td>
<td colspan="2" class="diff-lineno">Line 96:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>| 11.4 TOPS (int8)</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>| 11.4 TOPS (int8)</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|-</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|-</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>| 5.7 <del class="diffchange diffchange-inline">FLOPS </del>(FP16)</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>| 5.7 <ins class="diffchange diffchange-inline">TFLOPS </ins>(FP16)</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|}</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
</table>
141.113.3.11
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=83857&oldid=prev
72.31.25.172: Corrected errors based on latest NVIDIA specs
2018-11-05T03:09:52Z
<p>Corrected errors based on latest NVIDIA specs</p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 03:09, 5 November 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, particularly improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that <del class="diffchange diffchange-inline">they added int8 support for lower precision to the CUDA </del>tensor cores <del class="diffchange diffchange-inline">and those operate at the full 2x throughput </del>of <del class="diffchange diffchange-inline">the </del>FP16 <del class="diffchange diffchange-inline">[[FLOPS]]. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672)</del>. All of this yields 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, particularly improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that <ins class="diffchange diffchange-inline">each Volta multiprocessor contains eight </ins>tensor cores <ins class="diffchange diffchange-inline">each </ins>of <ins class="diffchange diffchange-inline">which can perform 16x </ins>FP16 <ins class="diffchange diffchange-inline">MACs or 32x INT8 MACs per cycle</ins>. All of this yields <ins class="diffchange diffchange-inline">a maximum </ins>22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
72.31.25.172
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82988&oldid=prev
David: /* GPU */
2018-09-29T14:42:47Z
<p><span dir="auto"><span class="autocomment">GPU</span></span></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 14:42, 29 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, <del class="diffchange diffchange-inline">paticlarly </del>improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 [[FLOPS]]. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, <ins class="diffchange diffchange-inline">particularly </ins>improving inference performance over training. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 [[FLOPS]]. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
David
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82987&oldid=prev
David: /* GPU */
2018-09-29T14:42:31Z
<p><span dir="auto"><span class="autocomment">GPU</span></span></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 14:42, 29 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, paticlarly <del class="diffchange diffchange-inline">adding </del>inference performance <del class="diffchange diffchange-inline">overtraining</del>. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 [[FLOPS]]. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, paticlarly <ins class="diffchange diffchange-inline">improving </ins>inference performance <ins class="diffchange diffchange-inline">over training</ins>. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 [[FLOPS]]. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
David
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82701&oldid=prev
David at 15:20, 22 September 2018
2018-09-22T15:20:13Z
<p></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 15:20, 22 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, paticlarly adding inference performance overtraining. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 FLOPS. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine learning market, paticlarly adding inference performance overtraining. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 <ins class="diffchange diffchange-inline">[[</ins>FLOPS<ins class="diffchange diffchange-inline">]]</ins>. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
<tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l88" >Line 88:</td>
<td colspan="2" class="diff-lineno">Line 88:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>==== Deep Learning Accelerator ====</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>==== Deep Learning Accelerator ====</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>The other accelerators on-die is the deep learning accelerator (DLA) which is actually a physical implementation of the open source Nvidia NVDLA architecture. Xavier has two instances of NVDLA which can offer a peak theoretical performance of 5.7 teraFLOPS (half precision FP) or twice the throughput at 11.4 TOPS for int8.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>The other accelerators on-die is the deep learning accelerator (DLA) which is actually a physical implementation of the open source Nvidia NVDLA architecture. Xavier has two instances of NVDLA which can offer a peak theoretical performance of 5.7 <ins class="diffchange diffchange-inline">[[</ins>teraFLOPS<ins class="diffchange diffchange-inline">]] </ins>(half precision FP) or twice the throughput at 11.4 TOPS for int8.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
David
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82428&oldid=prev
4.14.180.238: earning->learning
2018-09-14T11:11:58Z
<p>earning->learning</p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 11:11, 14 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l48" >Line 48:</td>
<td colspan="2" class="diff-lineno">Line 48:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>[[File:xavier gpu.svg|right|thumb|300px|GPU Block Diagram]]</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{main|nvidia/microarchitectures/volta|l1=Volta}}</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine <del class="diffchange diffchange-inline">earning </del>market, paticlarly adding inference performance overtraining. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 FLOPS. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine <ins class="diffchange diffchange-inline">learning </ins>market, paticlarly adding inference performance overtraining. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 FLOPS. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{| class="wikitable"</div></td></tr>
</table>
4.14.180.238
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82267&oldid=prev
David at 20:24, 10 September 2018
2018-09-10T20:24:56Z
<p></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 20:24, 10 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l1" >Line 1:</td>
<td colspan="2" class="diff-lineno">Line 1:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{nvidia title|Tegra Xavier}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{nvidia title|Tegra Xavier}}</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{chip</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{chip</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|name=<del class="diffchange diffchange-inline">Tegra </del>Xavier</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|name=Xavier</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|image=xavier soc chip.png</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|image=xavier soc chip.png</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|designer=Nvidia</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|designer=Nvidia</div></td></tr>
</table>
David
https://en.wikichip.org/w/index.php?title=nvidia/tegra/xavier&diff=82266&oldid=prev
David at 20:24, 10 September 2018
2018-09-10T20:24:34Z
<p></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class='diff-marker' />
<col class='diff-content' />
<col class='diff-marker' />
<col class='diff-content' />
<tr style='vertical-align: top;' lang='en'>
<td colspan='2' style="background-color: white; color:black; text-align: center;">← Older revision</td>
<td colspan='2' style="background-color: white; color:black; text-align: center;">Revision as of 20:24, 10 September 2018</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l1" >Line 1:</td>
<td colspan="2" class="diff-lineno">Line 1:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{nvidia title|Tegra Xavier}}</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{nvidia title|Tegra Xavier}}</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{chip</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>{{chip</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>|name=Xavier</div></td><td class='diff-marker'>+</td><td style="color:black; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>|name=<ins class="diffchange diffchange-inline">Tegra </ins>Xavier</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|image=xavier soc chip.png</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|image=xavier soc chip.png</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|designer=Nvidia</div></td><td class='diff-marker'> </td><td style="background-color: #f9f9f9; color: #333333; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #e6e6e6; vertical-align: top; white-space: pre-wrap;"><div>|designer=Nvidia</div></td></tr>
</table>
David